phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-04 20:14:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: Adding OpenAI Compatible Prompts API

Browse source 
Signed-off-by: Francisco Javier Arceo <farceo@redhat.com>
This commit is contained in:
Francisco Javier Arceo 2025-09-03 14:14:54 -04:00
parent 30117dea22
commit 8b00883abd
181 changed files with 21356 additions and 10332 deletions
Download patch file Download diff file Expand all files Collapse all files

2
.gitignore vendored
View file

@ -26,5 +26,7 @@ venv/
pytest-report.xml
.coverage
.python-version
AGENTS.md
server.log
CLAUDE.md
.claude/

2
.pre-commit-config.yaml
View file

@ -86,7 +86,7 @@ repos:
language: python
pass_filenames: false
require_serial: true
files: ^llama_stack/templates/.*$|^llama_stack/providers/.*/inference/.*/models\.py$
files: ^llama_stack/distributions/.*$|^llama_stack/providers/.*/inference/.*/models\.py$
- id: provider-codegen
name: Provider Codegen
additional_dependencies:

423
docs/_static/llama-stack-spec.html vendored
View file

@ -633,6 +633,80 @@
}
}
},
"/v1/prompts": {
"get": {
"responses": {
"200": {
"description": "A ListPromptsResponse containing all prompts.",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/ListPromptsResponse"
}
}
}
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "List all prompts.",
"parameters": []
},
"post": {
"responses": {
"200": {
"description": "The created Prompt resource.",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/Prompt"
}
}
}
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "Create a new prompt.",
"parameters": [],
"requestBody": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/CreatePromptRequest"
}
}
},
"required": true
}
}
},
"/v1/agents/{agent_id}": {
"get": {
"responses": {
@ -901,6 +975,143 @@
]
}
},
"/v1/prompts/{prompt_id}": {
"get": {
"responses": {
"200": {
"description": "A Prompt resource.",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/Prompt"
}
}
}
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "Get a prompt by its identifier and optional version.",
"parameters": [
{
"name": "prompt_id",
"in": "path",
"description": "The identifier of the prompt to get.",
"required": true,
"schema": {
"type": "string"
}
},
{
"name": "version",
"in": "query",
"description": "The version of the prompt to get (defaults to latest).",
"required": false,
"schema": {
"type": "string"
}
}
]
},
"post": {
"responses": {
"200": {
"description": "The updated Prompt resource with incremented version.",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/Prompt"
}
}
}
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "Update an existing prompt (increments version).",
"parameters": [
{
"name": "prompt_id",
"in": "path",
"description": "The identifier of the prompt to update.",
"required": true,
"schema": {
"type": "string"
}
}
],
"requestBody": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/UpdatePromptRequest"
}
}
},
"required": true
}
},
"delete": {
"responses": {
"200": {
"description": "OK"
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "Delete a prompt.",
"parameters": [
{
"name": "prompt_id",
"in": "path",
"description": "The identifier of the prompt to delete.",
"required": true,
"schema": {
"type": "string"
}
}
]
}
},
"/v1/inference/embeddings": {
"post": {
"responses": {
@ -4129,7 +4340,7 @@
"tags": [
"Files"
],
"description": "Upload a file that can be used across various endpoints.\nThe file upload should be a multipart form request with:\n- file: The File object (not file name) to be uploaded.\n- purpose: The intended purpose of the uploaded file.",
"description": "Upload a file that can be used across various endpoints.\nThe file upload should be a multipart form request with:\n- file: The File object (not file name) to be uploaded.\n- purpose: The intended purpose of the uploaded file.\n- expires_after: Optional form values describing expiration for the file. Expected expires_after[anchor] = \"created_at\", expires_after[seconds] = <int>. Seconds must be between 3600 and 2592000 (1 hour to 30 days).",
"parameters": [],
"requestBody": {
"content": {
@ -4143,11 +4354,33 @@
},
"purpose": {
"$ref": "#/components/schemas/OpenAIFilePurpose"
},
"expires_after_anchor": {
"oneOf": [
{
"type": "string"
},
{
"type": "null"
}
]
},
"expires_after_seconds": {
"oneOf": [
{
"type": "integer"
},
{
"type": "null"
}
]
}
},
"required": [
"file",
"purpose"
"purpose",
"expires_after_anchor",
"expires_after_seconds"
]
}
}
@ -4985,6 +5218,59 @@
}
}
},
"/v1/prompts/{prompt_id}/default-version": {
"post": {
"responses": {
"200": {
"description": "The prompt with the specified version now set as default.",
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/Prompt"
}
}
}
},
"400": {
"$ref": "#/components/responses/BadRequest400"
},
"429": {
"$ref": "#/components/responses/TooManyRequests429"
},
"500": {
"$ref": "#/components/responses/InternalServerError500"
},
"default": {
"$ref": "#/components/responses/DefaultError"
}
},
"tags": [
"Prompts"
],
"description": "Set which version of a prompt should be the default in get_prompt (latest).",
"parameters": [
{
"name": "prompt_id",
"in": "path",
"description": "The identifier of the prompt.",
"required": true,
"schema": {
"type": "string"
}
}
],
"requestBody": {
"content": {
"application/json": {
"schema": {
"$ref": "#/components/schemas/SetDefaultVersionRequest"
}
}
},
"required": true
}
}
},
"/v1/post-training/supervised-fine-tune": {
"post": {
"responses": {
@ -9648,6 +9934,58 @@
],
"title": "OpenAIResponseObjectStreamResponseWebSearchCallSearching"
},
"CreatePromptRequest": {
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "The prompt text content with variable placeholders."
},
"variables": {
"type": "object",
"additionalProperties": {
"type": "string"
},
"description": "Dictionary of variable names to their default values."
}
},
"additionalProperties": false,
"required": [
"prompt"
],
"title": "CreatePromptRequest"
},
"Prompt": {
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "The system prompt text with variable placeholders. Variables are only supported when using the Responses API."
},
"version": {
"type": "string",
"description": "Version string (integer start at 1 cast as string, incremented on save)"
},
"prompt_id": {
"type": "string",
"description": "Unique identifier formatted as 'pmpt_<48-digit-hash>'"
},
"variables": {
"type": "object",
"additionalProperties": {
"type": "string"
},
"description": "Dictionary of prompt variable names and values"
}
},
"additionalProperties": false,
"required": [
"version",
"prompt_id"
],
"title": "Prompt",
"description": "A prompt resource representing a stored OpenAI Compatible prompt template in Llama Stack."
},
"OpenAIDeleteResponseObject": {
"type": "object",
"properties": {
@ -10274,7 +10612,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "benchmark",
"default": "benchmark",
@ -10901,7 +11240,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "dataset",
"default": "dataset",
@ -11051,7 +11391,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "model",
"default": "model",
@ -11316,7 +11657,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "scoring_function",
"default": "scoring_function",
@ -11424,7 +11766,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "shield",
"default": "shield",
@ -11669,7 +12012,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "tool",
"default": "tool",
@ -11751,7 +12095,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "tool_group",
"default": "tool_group",
@ -12045,7 +12390,8 @@
"scoring_function",
"benchmark",
"tool",
"tool_group"
"tool_group",
"prompt"
],
"const": "vector_db",
"default": "vector_db",
@ -12860,6 +13206,23 @@
"title": "OpenAIResponseObjectWithInput",
"description": "OpenAI response object extended with input context information."
},
"ListPromptsResponse": {
"type": "object",
"properties": {
"data": {
"type": "array",
"items": {
"$ref": "#/components/schemas/Prompt"
}
}
},
"additionalProperties": false,
"required": [
"data"
],
"title": "ListPromptsResponse",
"description": "Response model to list prompts."
},
"ListProvidersResponse": {
"type": "object",
"properties": {
@ -17106,6 +17469,20 @@
"title": "ScoreBatchResponse",
"description": "Response from batch scoring operations on datasets."
},
"SetDefaultVersionRequest": {
"type": "object",
"properties": {
"version": {
"type": "string",
"description": "The version to set as default."
}
},
"additionalProperties": false,
"required": [
"version"
],
"title": "SetDefaultVersionRequest"
},
"AlgorithmConfig": {
"oneOf": [
{
@ -17390,6 +17767,27 @@
"title": "SyntheticDataGenerationResponse",
"description": "Response from the synthetic data generation. Batch of (prompt, response, score) tuples that pass the threshold."
},
"UpdatePromptRequest": {
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "The updated prompt text content."
},
"variables": {
"type": "object",
"additionalProperties": {
"type": "string"
},
"description": "Updated dictionary of variable names to their default values."
}
},
"additionalProperties": false,
"required": [
"prompt"
],
"title": "UpdatePromptRequest"
},
"VersionInfo": {
"type": "object",
"properties": {
@ -17515,6 +17913,10 @@
{
"name": "PostTraining (Coming Soon)"
},
{
"name": "Prompts",
"x-displayName": "Protocol for prompt management operations."
},
{
"name": "Providers",
"x-displayName": "Providers API for inspecting, listing, and modifying providers and their configurations."
@ -17565,6 +17967,7 @@
"Inspect",
"Models",
"PostTraining (Coming Soon)",
"Prompts",
"Providers",
"Safety",
"Scoring",

297
docs/_static/llama-stack-spec.yaml vendored
View file

@ -427,6 +427,58 @@ paths:
schema:
$ref: '#/components/schemas/CreateOpenaiResponseRequest'
required: true
/v1/prompts:
get:
responses:
'200':
description: >-
A ListPromptsResponse containing all prompts.
content:
application/json:
schema:
$ref: '#/components/schemas/ListPromptsResponse'
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: List all prompts.
parameters: []
post:
responses:
'200':
description: The created Prompt resource.
content:
application/json:
schema:
$ref: '#/components/schemas/Prompt'
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: Create a new prompt.
parameters: []
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/CreatePromptRequest'
required: true
/v1/agents/{agent_id}:
get:
responses:
@ -616,6 +668,103 @@ paths:
required: true
schema:
type: string
/v1/prompts/{prompt_id}:
get:
responses:
'200':
description: A Prompt resource.
content:
application/json:
schema:
$ref: '#/components/schemas/Prompt'
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: >-
Get a prompt by its identifier and optional version.
parameters:
- name: prompt_id
in: path
description: The identifier of the prompt to get.
required: true
schema:
type: string
- name: version
in: query
description: >-
The version of the prompt to get (defaults to latest).
required: false
schema:
type: string
post:
responses:
'200':
description: >-
The updated Prompt resource with incremented version.
content:
application/json:
schema:
$ref: '#/components/schemas/Prompt'
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: >-
Update an existing prompt (increments version).
parameters:
- name: prompt_id
in: path
description: The identifier of the prompt to update.
required: true
schema:
type: string
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/UpdatePromptRequest'
required: true
delete:
responses:
'200':
description: OK
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: Delete a prompt.
parameters:
- name: prompt_id
in: path
description: The identifier of the prompt to delete.
required: true
schema:
type: string
/v1/inference/embeddings:
post:
responses:
@ -2933,6 +3082,10 @@ paths:
- file: The File object (not file name) to be uploaded.
- purpose: The intended purpose of the uploaded file.
- expires_after: Optional form values describing expiration for the file.
Expected expires_after[anchor] = "created_at", expires_after[seconds] = <int>.
Seconds must be between 3600 and 2592000 (1 hour to 30 days).
parameters: []
requestBody:
content:
@ -2945,9 +3098,19 @@ paths:
format: binary
purpose:
$ref: '#/components/schemas/OpenAIFilePurpose'
expires_after_anchor:
oneOf:
- type: string
- type: 'null'
expires_after_seconds:
oneOf:
- type: integer
- type: 'null'
required:
- file
- purpose
- expires_after_anchor
- expires_after_seconds
required: true
/v1/openai/v1/models:
get:
@ -3532,6 +3695,43 @@ paths:
schema:
$ref: '#/components/schemas/ScoreBatchRequest'
required: true
/v1/prompts/{prompt_id}/default-version:
post:
responses:
'200':
description: >-
The prompt with the specified version now set as default.
content:
application/json:
schema:
$ref: '#/components/schemas/Prompt'
'400':
$ref: '#/components/responses/BadRequest400'
'429':
$ref: >-
#/components/responses/TooManyRequests429
'500':
$ref: >-
#/components/responses/InternalServerError500
default:
$ref: '#/components/responses/DefaultError'
tags:
- Prompts
description: >-
Set which version of a prompt should be the default in get_prompt (latest).
parameters:
- name: prompt_id
in: path
description: The identifier of the prompt.
required: true
schema:
type: string
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/SetDefaultVersionRequest'
required: true
/v1/post-training/supervised-fine-tune:
post:
responses:
@ -7134,6 +7334,53 @@ components:
- type
title: >-
OpenAIResponseObjectStreamResponseWebSearchCallSearching
CreatePromptRequest:
type: object
properties:
prompt:
type: string
description: >-
The prompt text content with variable placeholders.
variables:
type: object
additionalProperties:
type: string
description: >-
Dictionary of variable names to their default values.
additionalProperties: false
required:
- prompt
title: CreatePromptRequest
Prompt:
type: object
properties:
prompt:
type: string
description: >-
The system prompt text with variable placeholders. Variables are only
supported when using the Responses API.
version:
type: string
description: >-
Version string (integer start at 1 cast as string, incremented on save)
prompt_id:
type: string
description: >-
Unique identifier formatted as 'pmpt_<48-digit-hash>'
variables:
type: object
additionalProperties:
type: string
description: >-
Dictionary of prompt variable names and values
additionalProperties: false
required:
- version
- prompt_id
title: Prompt
description: >-
A prompt resource representing a stored OpenAI Compatible prompt template
in Llama Stack.
OpenAIDeleteResponseObject:
type: object
properties:
@ -7607,6 +7854,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: benchmark
default: benchmark
description: The resource type, always benchmark
@ -8093,6 +8341,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: dataset
default: dataset
description: >-
@ -8205,6 +8454,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: model
default: model
description: >-
@ -8396,6 +8646,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: scoring_function
default: scoring_function
description: >-
@ -8472,6 +8723,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: shield
default: shield
description: The resource type, always shield
@ -8651,6 +8903,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: tool
default: tool
description: Type of resource, always 'tool'
@ -8709,6 +8962,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: tool_group
default: tool_group
description: Type of resource, always 'tool_group'
@ -8937,6 +9191,7 @@ components:
- benchmark
- tool
- tool_group
- prompt
const: vector_db
default: vector_db
description: >-
@ -9563,6 +9818,18 @@ components:
title: OpenAIResponseObjectWithInput
description: >-
OpenAI response object extended with input context information.
ListPromptsResponse:
type: object
properties:
data:
type: array
items:
$ref: '#/components/schemas/Prompt'
additionalProperties: false
required:
- data
title: ListPromptsResponse
description: Response model to list prompts.
ListProvidersResponse:
type: object
properties:
@ -12708,6 +12975,16 @@ components:
title: ScoreBatchResponse
description: >-
Response from batch scoring operations on datasets.
SetDefaultVersionRequest:
type: object
properties:
version:
type: string
description: The version to set as default.
additionalProperties: false
required:
- version
title: SetDefaultVersionRequest
AlgorithmConfig:
oneOf:
- $ref: '#/components/schemas/LoraFinetuningConfig'
@ -12904,6 +13181,22 @@ components:
description: >-
Response from the synthetic data generation. Batch of (prompt, response, score)
tuples that pass the threshold.
UpdatePromptRequest:
type: object
properties:
prompt:
type: string
description: The updated prompt text content.
variables:
type: object
additionalProperties:
type: string
description: >-
Updated dictionary of variable names to their default values.
additionalProperties: false
required:
- prompt
title: UpdatePromptRequest
VersionInfo:
type: object
properties:
@ -13015,6 +13308,9 @@ tags:
- name: Inspect
- name: Models
- name: PostTraining (Coming Soon)
- name: Prompts
x-displayName: >-
Protocol for prompt management operations.
- name: Providers
x-displayName: >-
Providers API for inspecting, listing, and modifying providers and their configurations.
@ -13042,6 +13338,7 @@ x-tagGroups:
- Inspect
- Models
- PostTraining (Coming Soon)
- Prompts
- Providers
- Safety
- Scoring

9
docs/source/contributing/testing/record-replay.md
View file

@ -40,18 +40,15 @@ The system patches OpenAI and Ollama client methods to intercept calls before th
### Storage Architecture
Recordings use a two-tier storage system optimized for both speed and debuggability:
Recordings are stored as JSON files in the recording directory. They are looked up by their request hash.
```
recordings/
├── index.sqlite # Fast lookup by request hash
└── responses/
├── abc123def456.json # Individual response files
└── def789ghi012.json
```
**SQLite index** enables O(log n) hash lookups and metadata queries without loading response bodies.
**JSON files** store complete request/response pairs in human-readable format for debugging.
## Recording Modes
@ -166,8 +163,8 @@ This preserves type safety - when replayed, you get the same Pydantic objects wi
Control recording behavior globally:
```bash
export LLAMA_STACK_TEST_INFERENCE_MODE=replay
export LLAMA_STACK_TEST_RECORDING_DIR=/path/to/recordings
export LLAMA_STACK_TEST_INFERENCE_MODE=replay # this is the default
export LLAMA_STACK_TEST_RECORDING_DIR=/path/to/recordings # default is tests/integration/recordings
pytest tests/integration/
```

1
docs/source/distributions/self_hosted_distro/nvidia.md
View file

@ -50,6 +50,7 @@ The following models are available by default:
- `meta/llama-3.2-11b-vision-instruct `
- `meta/llama-3.2-90b-vision-instruct `
- `meta/llama-3.3-70b-instruct `
- `nvidia/vila `
- `nvidia/llama-3.2-nv-embedqa-1b-v2 `
- `nvidia/nv-embedqa-e5-v5 `
- `nvidia/nv-embedqa-mistral-7b-v2 `

73
docs/source/providers/vector_io/remote_pgvector.md
View file

@ -12,6 +12,60 @@ That means you'll get fast and efficient vector retrieval.
- Easy to use
- Fully integrated with Llama Stack
There are three implementations of search for PGVectoIndex available:
1. Vector Search:
- How it works:
- Uses PostgreSQL's vector extension (pgvector) to perform similarity search
- Compares query embeddings against stored embeddings using Cosine distance or other distance metrics
- Eg. SQL query: SELECT document, embedding <=> %s::vector AS distance FROM table ORDER BY distance
-Characteristics:
- Semantic understanding - finds documents similar in meaning even if they don't share keywords
- Works with high-dimensional vector embeddings (typically 768, 1024, or higher dimensions)
- Best for: Finding conceptually related content, handling synonyms, cross-language search
2. Keyword Search
- How it works:
- Uses PostgreSQL's full-text search capabilities with tsvector and ts_rank
- Converts text to searchable tokens using to_tsvector('english', text). Default language is English.
- Eg. SQL query: SELECT document, ts_rank(tokenized_content, plainto_tsquery('english', %s)) AS score
- Characteristics:
- Lexical matching - finds exact keyword matches and variations
- Uses GIN (Generalized Inverted Index) for fast text search performance
- Scoring: Uses PostgreSQL's ts_rank function for relevance scoring
- Best for: Exact term matching, proper names, technical terms, Boolean-style queries
3. Hybrid Search
- How it works:
- Combines both vector and keyword search results
- Runs both searches independently, then merges results using configurable reranking
- Two reranking strategies available:
- Reciprocal Rank Fusion (RRF) - (default: 60.0)
- Weighted Average - (default: 0.5)
- Characteristics:
- Best of both worlds: semantic understanding + exact matching
- Documents appearing in both searches get boosted scores
- Configurable balance between semantic and lexical matching
- Best for: General-purpose search where you want both precision and recall
4. Database Schema
The PGVector implementation stores data optimized for all three search types:
CREATE TABLE vector_store_xxx (
id TEXT PRIMARY KEY,
document JSONB, -- Original document
embedding vector(dimension), -- For vector search
content_text TEXT, -- Raw text content
tokenized_content TSVECTOR -- For keyword search
);
-- Indexes for performance
CREATE INDEX content_gin_idx ON table USING GIN(tokenized_content); -- Keyword search
-- Vector index created automatically by pgvector
## Usage
To use PGVector in your Llama Stack project, follow these steps:
@ -20,6 +74,25 @@ To use PGVector in your Llama Stack project, follow these steps:
2. Configure your Llama Stack project to use pgvector. (e.g. remote::pgvector).
3. Start storing and querying vectors.
## This is an example how you can set up your environment for using PGVector
1. Export env vars:
```bash
export ENABLE_PGVECTOR=true
export PGVECTOR_HOST=localhost
export PGVECTOR_PORT=5432
export PGVECTOR_DB=llamastack
export PGVECTOR_USER=llamastack
export PGVECTOR_PASSWORD=llamastack
```
2. Create DB:
```bash
psql -h localhost -U postgres -c "CREATE ROLE llamastack LOGIN PASSWORD 'llamastack';"
psql -h localhost -U postgres -c "CREATE DATABASE llamastack OWNER llamastack;"
psql -h localhost -U llamastack -d llamastack -c "CREATE EXTENSION IF NOT EXISTS vector;"
```
## Installation
You can install PGVector using docker:

1
docs/source/providers/vector_io/remote_weaviate.md
View file

@ -17,6 +17,7 @@ Weaviate supports:
- Metadata filtering
- Multi-modal retrieval
## Usage
To use Weaviate in your Llama Stack project, follow these steps:

1
docs/source/references/llama_stack_client_cli_reference.md
View file

@ -478,7 +478,6 @@ llama-stack-client scoring_functions list
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┓
┃ identifier ┃ provider_id ┃ description ┃ type ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━┩
│ basic::bfcl │ basic │ BFCL complex scoring │ scoring_function │
│ basic::docvqa │ basic │ DocVQA Visual Question & Answer scoring function │ scoring_function │
│ basic::equality │ basic │ Returns 1.0 if the input is equal to the target, 0.0 │ scoring_function │
│ │ │ otherwise. │ │

2
llama_stack/apis/datatypes.py
View file

@ -102,6 +102,7 @@ class Api(Enum, metaclass=DynamicApiMeta):
:cvar benchmarks: Benchmark suite management
:cvar tool_groups: Tool group organization
:cvar files: File storage and management
:cvar prompts: Prompt versions and management
:cvar inspect: Built-in system inspection and introspection
"""
@ -127,6 +128,7 @@ class Api(Enum, metaclass=DynamicApiMeta):
benchmarks = "benchmarks"
tool_groups = "tool_groups"
files = "files"
prompts = "prompts"
# built-in API
inspect = "inspect"

25
llama_stack/apis/files/files.py
View file

@ -5,10 +5,10 @@
# the root directory of this source tree.
from enum import StrEnum
from typing import Annotated, Literal, Protocol, runtime_checkable
from typing import Annotated, ClassVar, Literal, Protocol, runtime_checkable
from fastapi import File, Form, Response, UploadFile
from pydantic import BaseModel
from pydantic import BaseModel, Field
from llama_stack.apis.common.responses import Order
from llama_stack.providers.utils.telemetry.trace_protocol import trace_protocol
@ -49,6 +49,23 @@ class OpenAIFileObject(BaseModel):
purpose: OpenAIFilePurpose
@json_schema_type
class ExpiresAfter(BaseModel):
"""
Control expiration of uploaded files.
Params:
- anchor, must be "created_at"
- seconds, must be int between 3600 and 2592000 (1 hour to 30 days)
"""
MIN: ClassVar[int] = 3600 # 1 hour
MAX: ClassVar[int] = 2592000 # 30 days
anchor: Literal["created_at"]
seconds: int = Field(..., ge=3600, le=2592000)
@json_schema_type
class ListOpenAIFileResponse(BaseModel):
"""
@ -92,6 +109,9 @@ class Files(Protocol):
self,
file: Annotated[UploadFile, File()],
purpose: Annotated[OpenAIFilePurpose, Form()],
expires_after_anchor: Annotated[str | None, Form(alias="expires_after[anchor]")] = None,
expires_after_seconds: Annotated[int | None, Form(alias="expires_after[seconds]")] = None,
# TODO: expires_after is producing strange openapi spec, params are showing up as a required w/ oneOf being null
) -> OpenAIFileObject:
"""
Upload a file that can be used across various endpoints.
@ -99,6 +119,7 @@ class Files(Protocol):
The file upload should be a multipart form request with:
- file: The File object (not file name) to be uploaded.
- purpose: The intended purpose of the uploaded file.
- expires_after: Optional form values describing expiration for the file. Expected expires_after[anchor] = "created_at", expires_after[seconds] = <int>. Seconds must be between 3600 and 2592000 (1 hour to 30 days).
:param file: The uploaded file object containing content and metadata (filename, content_type, etc.).
:param purpose: The intended purpose of the uploaded file (e.g., "assistants", "fine-tune").

9
llama_stack/apis/prompts/__init__.py Normal file
View file

@ -0,0 +1,9 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from .prompts import CreatePromptRequest, ListPromptsResponse, Prompt, Prompts, UpdatePromptRequest
__all__ = ["Prompt", "Prompts", "ListPromptsResponse", "CreatePromptRequest", "UpdatePromptRequest"]

173
llama_stack/apis/prompts/prompts.py Normal file
View file

@ -0,0 +1,173 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import secrets
from typing import Protocol, runtime_checkable
from pydantic import BaseModel, Field, field_validator
from llama_stack.providers.utils.telemetry.trace_protocol import trace_protocol
from llama_stack.schema_utils import json_schema_type, webmethod
@json_schema_type
class Prompt(BaseModel):
"""A prompt resource representing a stored OpenAI Compatible prompt template in Llama Stack.
:param prompt: The system prompt text with variable placeholders. Variables are only supported when using the Responses API.
:param version: Version string (integer start at 1 cast as string, incremented on save)
:param prompt_id: Unique identifier formatted as 'pmpt_<48-digit-hash>'
:param variables: Dictionary of prompt variable names and values
"""
prompt: str | None = Field(default=None, description="The system prompt with variable placeholders")
version: str = Field(description="Version string (integer start at 1 cast as string)")
prompt_id: str = Field(description="Unique identifier in format 'pmpt_<48-digit-hash>'")
variables: dict[str, str] | None = Field(
default_factory=dict, description="Variables for dynamic injection using {{variable}} syntax"
)
@field_validator("prompt_id")
@classmethod
def validate_prompt_id(cls, prompt_id: str) -> str:
if not isinstance(prompt_id, str):
raise TypeError("prompt_id must be a string in format 'pmpt_<48-digit-hash>'")
if not prompt_id.startswith("pmpt_"):
raise ValueError("prompt_id must start with 'pmpt_' prefix")
hex_part = prompt_id[5:]
if len(hex_part) != 48:
raise ValueError("prompt_id must be in format 'pmpt_<48-digit-hash>' (48 lowercase hex chars)")
for char in hex_part:
if char not in "0123456789abcdef":
raise ValueError("prompt_id hex part must contain only lowercase hex characters [0-9a-f]")
return prompt_id
@field_validator("version")
@classmethod
def validate_version(cls, prompt_version: str) -> str:
try:
int_version = int(prompt_version)
if int_version < 1:
raise ValueError("version must be >= 1")
except ValueError as e:
if "invalid literal" in str(e):
raise ValueError("version must be a string representation of an integer") from e
raise
return prompt_version
@classmethod
def generate_prompt_id(cls) -> str:
# Generate 48 hex characters (24 bytes)
random_bytes = secrets.token_bytes(24)
hex_string = random_bytes.hex()
return f"pmpt_{hex_string}"
class CreatePromptRequest(BaseModel):
"""Request model to create a prompt."""
prompt: str = Field(description="The prompt text content")
variables: dict[str, str] = Field(default_factory=dict, description="Variables for dynamic injection")
class UpdatePromptRequest(BaseModel):
"""Request model for updating a prompt."""
prompt: str = Field(description="The prompt text content")
variables: dict[str, str] = Field(default_factory=dict, description="Variables for dynamic injection")
class ListPromptsResponse(BaseModel):
"""Response model to list prompts."""
data: list[Prompt]
@runtime_checkable
@trace_protocol
class Prompts(Protocol):
"""Protocol for prompt management operations."""
@webmethod(route="/prompts", method="GET")
async def list_prompts(self) -> ListPromptsResponse:
"""List all prompts.
:returns: A ListPromptsResponse containing all prompts.
"""
...
@webmethod(route="/prompts/{prompt_id:path}", method="GET")
async def get_prompt(
self,
prompt_id: str,
version: str | None = None,
) -> Prompt:
"""Get a prompt by its identifier and optional version.
:param prompt_id: The identifier of the prompt to get.
:param version: The version of the prompt to get (defaults to latest).
:returns: A Prompt resource.
"""
...
@webmethod(route="/prompts", method="POST")
async def create_prompt(
self,
prompt: str,
variables: dict[str, str] | None = None,
) -> Prompt:
"""Create a new prompt.
:param prompt: The prompt text content with variable placeholders.
:param variables: Dictionary of variable names to their default values.
:returns: The created Prompt resource.
"""
...
@webmethod(route="/prompts/{prompt_id:path}", method="PUT")
async def update_prompt(
self,
prompt_id: str,
prompt: str,
variables: dict[str, str] | None = None,
) -> Prompt:
"""Update an existing prompt (increments version).
:param prompt_id: The identifier of the prompt to update.
:param prompt: The updated prompt text content.
:param variables: Updated dictionary of variable names to their default values.
:returns: The updated Prompt resource with incremented version.
"""
...
@webmethod(route="/prompts/{prompt_id:path}", method="DELETE")
async def delete_prompt(
self,
prompt_id: str,
) -> None:
"""Delete a prompt.
:param prompt_id: The identifier of the prompt to delete.
"""
...
@webmethod(route="/prompts/{prompt_id:path}/default-version", method="PUT")
async def set_default_version(
self,
prompt_id: str,
version: str,
) -> Prompt:
"""Set which version of a prompt should be the default in get_prompt (latest).
:param prompt_id: The identifier of the prompt.
:param version: The version to set as default.
:returns: The prompt with the specified version now set as default.
"""
...

1
llama_stack/apis/resource.py
View file

@ -19,6 +19,7 @@ class ResourceType(StrEnum):
benchmark = "benchmark"
tool = "tool"
tool_group = "tool_group"
prompt = "prompt"
class Resource(BaseModel):

0
llama_stack/providers/inline/scoring/basic/utils/bfcl/__init__.py → llama_stack/core/prompts/__init__.py
View file

182
llama_stack/core/prompts/prompts.py Normal file
View file

@ -0,0 +1,182 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import json
from typing import Any
from pydantic import BaseModel
from llama_stack.apis.prompts import ListPromptsResponse, Prompt, Prompts
from llama_stack.providers.utils.kvstore import KVStore, kvstore_impl
from llama_stack.providers.utils.kvstore.config import KVStoreConfig
class PromptServiceConfig(BaseModel):
"""Configuration for the built-in prompt service.
:param kvstore: Configuration for the key-value store backend
"""
kvstore: KVStoreConfig
async def get_provider_impl(config: PromptServiceConfig, deps: dict[Any, Any]):
"""Get the prompt service implementation."""
impl = PromptServiceImpl(config, deps)
await impl.initialize()
return impl
class PromptServiceImpl(Prompts):
"""Built-in prompt service implementation using KVStore."""
def __init__(self, config: PromptServiceConfig, deps: dict[Any, Any]):
self.config = config
self.deps = deps
self.kvstore: KVStore
async def initialize(self) -> None:
self.kvstore = await kvstore_impl(self.config.kvstore)
def _get_prompt_key(self, prompt_id: str, version: str | None = None) -> str:
if version:
return f"prompts:v1:{prompt_id}:{version}"
return f"prompts:v1:{prompt_id}:default"
def _get_version_key(self, prompt_id: str, version: str) -> str:
"""Get the KVStore key for a specific prompt version."""
return f"prompts:v1:{prompt_id}:{version}"
def _get_list_key_prefix(self) -> str:
"""Get the key prefix for listing prompts."""
return "prompts:v1:"
def _serialize_prompt(self, prompt: Prompt) -> str:
"""Serialize a prompt to JSON string for storage."""
return json.dumps(
{
"prompt_id": prompt.prompt_id,
"prompt": prompt.prompt,
"version": prompt.version,
"variables": prompt.variables or {},
}
)
def _deserialize_prompt(self, data: str) -> Prompt:
"""Deserialize a prompt from JSON string."""
obj = json.loads(data)
return Prompt(
prompt_id=obj["prompt_id"], prompt=obj["prompt"], version=obj["version"], variables=obj.get("variables", {})
)
async def list_prompts(self) -> ListPromptsResponse:
"""List all prompts (default versions only)."""
prefix = self._get_list_key_prefix()
keys = await self.kvstore.keys_in_range(prefix, prefix + "\xff")
prompts = []
for key in keys:
if key.endswith(":default"):
try:
default_version = await self.kvstore.get(key)
if default_version:
prompt_id = key.replace(prefix, "").replace(":default", "")
version_key = self._get_version_key(prompt_id, default_version)
data = await self.kvstore.get(version_key)
if data:
prompt = self._deserialize_prompt(data)
prompts.append(prompt)
except (json.JSONDecodeError, KeyError):
continue
prompts.sort(key=lambda p: p.prompt_id or "", reverse=True)
return ListPromptsResponse(data=prompts)
async def get_prompt(self, prompt_id: str, version: str | None = None) -> Prompt:
"""Get a prompt by its identifier and optional version."""
if version:
key = self._get_version_key(prompt_id, version)
data = await self.kvstore.get(key)
if data is None:
raise ValueError(f"Prompt {prompt_id} version {version} not found")
else:
default_key = self._get_prompt_key(prompt_id)
default_version = await self.kvstore.get(default_key)
if default_version is None:
raise ValueError(f"Prompt with ID '{prompt_id}' not found")
key = self._get_version_key(prompt_id, default_version)
data = await self.kvstore.get(key)
if data is None:
raise ValueError(f"Prompt with ID '{prompt_id}' not found")
return self._deserialize_prompt(data)
async def create_prompt(
self,
prompt: str,
variables: dict[str, str] | None = None,
) -> Prompt:
"""Create a new prompt."""
if variables is None:
variables = {}
prompt_obj = Prompt(prompt_id=Prompt.generate_prompt_id(), prompt=prompt, version="1", variables=variables)
version_key = self._get_version_key(prompt_obj.prompt_id, "1")
data = self._serialize_prompt(prompt_obj)
await self.kvstore.set(version_key, data)
default_key = self._get_prompt_key(prompt_obj.prompt_id)
await self.kvstore.set(default_key, "1")
return prompt_obj
async def update_prompt(
self,
prompt_id: str,
prompt: str,
variables: dict[str, str] | None = None,
) -> Prompt:
"""Update an existing prompt (increments version)."""
if variables is None:
variables = {}
current_prompt = await self.get_prompt(prompt_id)
new_version = str(int(current_prompt.version) + 1)
updated_prompt = Prompt(prompt_id=prompt_id, prompt=prompt, version=new_version, variables=variables)
version_key = self._get_version_key(prompt_id, new_version)
data = self._serialize_prompt(updated_prompt)
await self.kvstore.set(version_key, data)
default_key = self._get_prompt_key(prompt_id)
await self.kvstore.set(default_key, new_version)
return updated_prompt
async def delete_prompt(self, prompt_id: str) -> None:
"""Delete a prompt and all its versions."""
await self.get_prompt(prompt_id)
prefix = f"prompts:v1:{prompt_id}:"
keys = await self.kvstore.keys_in_range(prefix, prefix + "\xff")
for key in keys:
await self.kvstore.delete(key)
async def set_default_version(self, prompt_id: str, version: str) -> Prompt:
"""Set which version of a prompt should be the default (latest)."""
version_key = self._get_version_key(prompt_id, version)
data = await self.kvstore.get(version_key)
if data is None:
raise ValueError(f"Prompt {prompt_id} version {version} not found")
default_key = self._get_prompt_key(prompt_id)
await self.kvstore.set(default_key, version)
return self._deserialize_prompt(data)

10
llama_stack/core/resolver.py
View file

@ -19,6 +19,7 @@ from llama_stack.apis.inference import Inference, InferenceProvider
from llama_stack.apis.inspect import Inspect
from llama_stack.apis.models import Models
from llama_stack.apis.post_training import PostTraining
from llama_stack.apis.prompts import Prompts
from llama_stack.apis.providers import Providers as ProvidersAPI
from llama_stack.apis.safety import Safety
from llama_stack.apis.scoring import Scoring
@ -93,6 +94,7 @@ def api_protocol_map(external_apis: dict[Api, ExternalApiSpec] | None = None) ->
Api.tool_groups: ToolGroups,
Api.tool_runtime: ToolRuntime,
Api.files: Files,
Api.prompts: Prompts,
}
if external_apis:
@ -284,7 +286,15 @@ async def instantiate_providers(
if provider.provider_id is None:
continue
try:
deps = {a: impls[a] for a in provider.spec.api_dependencies}
except KeyError as e:
missing_api = e.args[0]
raise RuntimeError(
f"Failed to resolve '{provider.spec.api.value}' provider '{provider.provider_id}' of type '{provider.spec.provider_type}': "
f"required dependency '{missing_api.value}' is not available. "
f"Please add a '{missing_api.value}' provider to your configuration or check if the provider is properly configured."
) from e
for a in provider.spec.optional_api_dependencies:
if a in impls:
deps[a] = impls[a]

2
llama_stack/core/routers/inference.py
View file

@ -755,7 +755,7 @@ class InferenceRouter(Inference):
choices_data[idx] = {
"content_parts": [],
"tool_calls_builder": {},
"finish_reason": None,
"finish_reason": "stop",
"logprobs_content_parts": [],
}
current_choice_data = choices_data[idx]

5
llama_stack/core/server/server.py
View file

@ -132,9 +132,9 @@ def translate_exception(exc: Exception) -> HTTPException | RequestValidationErro
},
)
elif isinstance(exc, ConflictError):
return HTTPException(status_code=409, detail=str(exc))
return HTTPException(status_code=httpx.codes.CONFLICT, detail=str(exc))
elif isinstance(exc, ResourceNotFoundError):
return HTTPException(status_code=404, detail=str(exc))
return HTTPException(status_code=httpx.codes.NOT_FOUND, detail=str(exc))
elif isinstance(exc, ValueError):
return HTTPException(status_code=httpx.codes.BAD_REQUEST, detail=f"Invalid value: {str(exc)}")
elif isinstance(exc, BadRequestError):
@ -513,6 +513,7 @@ def main(args: argparse.Namespace | None = None):
apis_to_serve.add("inspect")
apis_to_serve.add("providers")
apis_to_serve.add("prompts")
for api_str in apis_to_serve:
api = Api(api_str)

19
llama_stack/core/stack.py
View file

@ -24,6 +24,7 @@ from llama_stack.apis.inference import Inference
from llama_stack.apis.inspect import Inspect
from llama_stack.apis.models import Models
from llama_stack.apis.post_training import PostTraining
from llama_stack.apis.prompts import Prompts
from llama_stack.apis.providers import Providers
from llama_stack.apis.safety import Safety
from llama_stack.apis.scoring import Scoring
@ -37,6 +38,7 @@ from llama_stack.apis.vector_io import VectorIO
from llama_stack.core.datatypes import Provider, StackRunConfig
from llama_stack.core.distribution import get_provider_registry
from llama_stack.core.inspect import DistributionInspectConfig, DistributionInspectImpl
from llama_stack.core.prompts.prompts import PromptServiceConfig, PromptServiceImpl
from llama_stack.core.providers import ProviderImpl, ProviderImplConfig
from llama_stack.core.resolver import ProviderRegistry, resolve_impls
from llama_stack.core.routing_tables.common import CommonRoutingTableImpl
@ -44,6 +46,7 @@ from llama_stack.core.store.registry import create_dist_registry
from llama_stack.core.utils.dynamic import instantiate_class_type
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import Api
from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
logger = get_logger(name=__name__, category="core")
@ -72,6 +75,7 @@ class LlamaStack(
ToolRuntime,
RAGToolRuntime,
Files,
Prompts,
):
pass
@ -105,12 +109,12 @@ async def register_resources(run_config: StackRunConfig, impls: dict[Api, Any]):
method = getattr(impls[api], register_method)
for obj in objects:
logger.debug(f"registering {rsrc.capitalize()} {obj} for provider {obj.provider_id}")
if hasattr(obj, "provider_id"):
# Do not register models on disabled providers
if hasattr(obj, "provider_id") and (not obj.provider_id or obj.provider_id == "__disabled__"):
if not obj.provider_id or obj.provider_id == "__disabled__":
logger.debug(f"Skipping {rsrc.capitalize()} registration for disabled provider.")
continue
logger.debug(f"registering {rsrc.capitalize()} {obj} for provider {obj.provider_id}")
# we want to maintain the type information in arguments to method.
# instead of method(**obj.model_dump()), which may convert a typed attr to a dict,
@ -305,6 +309,12 @@ def add_internal_implementations(impls: dict[Api, Any], run_config: StackRunConf
)
impls[Api.providers] = providers_impl
prompts_impl = PromptServiceImpl(
PromptServiceConfig(kvstore=SqliteKVStoreConfig(db_path=os.path.expanduser("~/.llama-stack/prompts.db"))),
deps=impls,
)
impls[Api.prompts] = prompts_impl
# Produces a stack of providers for the given run config. Not all APIs may be
# asked for in the run config.
@ -329,6 +339,9 @@ async def construct_stack(
# Add internal implementations after all other providers are resolved
add_internal_implementations(impls, run_config)
if Api.prompts in impls:
await impls[Api.prompts].initialize()
await register_resources(run_config, impls)
await refresh_registry_once(impls)

5
llama_stack/distributions/nvidia/run.yaml
View file

@ -134,6 +134,11 @@ models:
provider_id: nvidia
provider_model_id: meta/llama-3.3-70b-instruct
model_type: llm
- metadata: {}
model_id: nvidia/vila
provider_id: nvidia
provider_model_id: nvidia/vila
model_type: llm
- metadata:
embedding_dimension: 2048
context_length: 8192

16
llama_stack/distributions/open-benchmark/open_benchmark.py
View file

@ -43,7 +43,7 @@ def get_inference_providers() -> tuple[list[Provider], dict[str, list[ProviderMo
"openai",
[
ProviderModelEntry(
provider_model_id="openai/gpt-4o",
provider_model_id="gpt-4o",
model_type=ModelType.llm,
)
],
@ -53,7 +53,7 @@ def get_inference_providers() -> tuple[list[Provider], dict[str, list[ProviderMo
"anthropic",
[
ProviderModelEntry(
provider_model_id="anthropic/claude-3-5-sonnet-latest",
provider_model_id="claude-3-5-sonnet-latest",
model_type=ModelType.llm,
)
],
@ -206,13 +206,6 @@ def get_distribution_template() -> DistributionTemplate:
uri="huggingface://datasets/llamastack/math_500?split=test",
),
),
DatasetInput(
dataset_id="bfcl",
purpose=DatasetPurpose.eval_messages_answer,
source=URIDataSource(
uri="huggingface://datasets/llamastack/bfcl_v3?split=train",
),
),
DatasetInput(
dataset_id="ifeval",
purpose=DatasetPurpose.eval_messages_answer,
@ -250,11 +243,6 @@ def get_distribution_template() -> DistributionTemplate:
dataset_id="math_500",
scoring_functions=["basic::regex_parser_math_response"],
),
BenchmarkInput(
benchmark_id="meta-reference-bfcl",
dataset_id="bfcl",
scoring_functions=["basic::bfcl"],
),
BenchmarkInput(
benchmark_id="meta-reference-ifeval",
dataset_id="ifeval",

19
llama_stack/distributions/open-benchmark/run.yaml
View file

@ -136,14 +136,14 @@ inference_store:
db_path: ${env.SQLITE_STORE_DIR:=~/.llama/distributions/open-benchmark}/inference_store.db
models:
- metadata: {}
model_id: openai/gpt-4o
model_id: gpt-4o
provider_id: openai
provider_model_id: openai/gpt-4o
provider_model_id: gpt-4o
model_type: llm
- metadata: {}
model_id: anthropic/claude-3-5-sonnet-latest
model_id: claude-3-5-sonnet-latest
provider_id: anthropic
provider_model_id: anthropic/claude-3-5-sonnet-latest
provider_model_id: claude-3-5-sonnet-latest
model_type: llm
- metadata: {}
model_id: gemini/gemini-1.5-flash
@ -188,12 +188,6 @@ datasets:
uri: huggingface://datasets/llamastack/math_500?split=test
metadata: {}
dataset_id: math_500
- purpose: eval/messages-answer
source:
type: uri
uri: huggingface://datasets/llamastack/bfcl_v3?split=train
metadata: {}
dataset_id: bfcl
- purpose: eval/messages-answer
source:
type: uri
@ -228,11 +222,6 @@ benchmarks:
- basic::regex_parser_math_response
metadata: {}
benchmark_id: meta-reference-math-500
- dataset_id: bfcl
scoring_functions:
- basic::bfcl
metadata: {}
benchmark_id: meta-reference-bfcl
- dataset_id: ifeval
scoring_functions:
- basic::ifeval

5
llama_stack/providers/inline/files/localfs/files.py
View file

@ -86,11 +86,16 @@ class LocalfsFilesImpl(Files):
self,
file: Annotated[UploadFile, File()],
purpose: Annotated[OpenAIFilePurpose, Form()],
expires_after_anchor: Annotated[str | None, Form(alias="expires_after[anchor]")] = None,
expires_after_seconds: Annotated[int | None, Form(alias="expires_after[seconds]")] = None,
) -> OpenAIFileObject:
"""Upload a file that can be used across various endpoints."""
if not self.sql_store:
raise RuntimeError("Files provider not initialized")
if expires_after_anchor is not None or expires_after_seconds is not None:
raise NotImplementedError("File expiration is not supported by this provider")
file_id = self._generate_file_id()
file_path = self._get_file_path(file_id)

2
llama_stack/providers/inline/scoring/basic/scoring.py
View file

@ -22,7 +22,6 @@ from llama_stack.providers.utils.common.data_schema_validator import (
)
from .config import BasicScoringConfig
from .scoring_fn.bfcl_scoring_fn import BFCLScoringFn
from .scoring_fn.docvqa_scoring_fn import DocVQAScoringFn
from .scoring_fn.equality_scoring_fn import EqualityScoringFn
from .scoring_fn.ifeval_scoring_fn import IfEvalScoringFn
@ -37,7 +36,6 @@ FIXED_FNS = [
SubsetOfScoringFn,
RegexParserScoringFn,
RegexParserMathResponseScoringFn,
BFCLScoringFn,
IfEvalScoringFn,
DocVQAScoringFn,
]

93
llama_stack/providers/inline/scoring/basic/scoring_fn/bfcl_scoring_fn.py
View file

@ -1,93 +0,0 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import json
import re
from typing import Any
from llama_stack.apis.scoring import ScoringResultRow
from llama_stack.apis.scoring_functions import ScoringFnParams
from llama_stack.providers.utils.scoring.base_scoring_fn import RegisteredBaseScoringFn
from ..utils.bfcl.ast_parser import decode_ast
from ..utils.bfcl.checker import ast_checker, is_empty_output
from .fn_defs.bfcl import bfcl
def postprocess(x: dict[str, Any], test_category: str) -> dict[str, Any]:
contain_func_call = False
error = None
error_type = None
checker_result = {}
try:
prediction = decode_ast(x["generated_answer"], x["language"]) or ""
contain_func_call = True
# if not is_function_calling_format_output(prediction):
if is_empty_output(prediction):
contain_func_call = False
error = "Did not output in the specified format. Note: the model_result is wrapped in a string to ensure json serializability."
error_type = "ast_decoder:decoder_wrong_output_format"
else:
checker_result = ast_checker(
json.loads(x["function"]),
prediction,
json.loads(x["ground_truth"]),
x["language"],
test_category=test_category,
model_name="",
)
except Exception as e:
prediction = ""
error = f"Invalid syntax. Failed to decode AST. {str(e)}"
error_type = "ast_decoder:decoder_failed"
return {
"prediction": prediction,
"contain_func_call": contain_func_call,
"valid": checker_result.get("valid", False),
"error": error or checker_result.get("error", ""),
"error_type": error_type or checker_result.get("error_type", ""),
}
def gen_valid(x: dict[str, Any]) -> dict[str, float]:
return {"valid": x["valid"]}
def gen_relevance_acc(x: dict[str, Any]) -> dict[str, float]:
# This function serves for both relevance and irrelevance tests, which share the exact opposite logic.
# If `test_category` is "irrelevance", the model is expected to output no function call.
# No function call means either the AST decoding fails (a error message is generated) or the decoded AST does not contain any function call (such as a empty list, `[]`).
# If `test_category` is "relevance", the model is expected to output to a function call, and empty list doesn't count as a function call.
acc = not x["contain_func_call"] if "irrelevance" in x["id"] else x["contain_func_call"]
return {"valid": float(acc)}
class BFCLScoringFn(RegisteredBaseScoringFn):
"""
A scoring_fn for BFCL
"""
def __init__(self, *args, **kwargs) -> None:
super().__init__(*args, **kwargs)
self.supported_fn_defs_registry = {
bfcl.identifier: bfcl,
}
async def score_row(
self,
input_row: dict[str, Any],
scoring_fn_identifier: str | None = "bfcl",
scoring_params: ScoringFnParams | None = None,
) -> ScoringResultRow:
test_category = re.sub(r"_[0-9_-]+$", "", input_row["id"])
score_result = postprocess(input_row, test_category)
if test_category in {"irrelevance", "live_relevance", "live_irrelevance"}:
score = gen_relevance_acc(score_result)["valid"]
else:
score = gen_valid(score_result)["valid"]
return {
"score": float(score),
}

21
llama_stack/providers/inline/scoring/basic/scoring_fn/fn_defs/bfcl.py
View file

@ -1,21 +0,0 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
from llama_stack.apis.common.type_system import NumberType
from llama_stack.apis.scoring_functions import (
AggregationFunctionType,
BasicScoringFnParams,
ScoringFn,
)
bfcl = ScoringFn(
identifier="basic::bfcl",
description="BFCL complex scoring",
return_type=NumberType(),
provider_id="basic",
provider_resource_id="bfcl",
params=BasicScoringFnParams(aggregation_functions=[AggregationFunctionType.accuracy]),
)

296
llama_stack/providers/inline/scoring/basic/utils/bfcl/ast_parser.py
View file

@ -1,296 +0,0 @@
# ruff: noqa
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import ast
from .tree_sitter import get_parser
def parse_java_function_call(source_code):
if not source_code.endswith(";"):
source_code += ";" # Necessary for the parser not to register an error
parser = get_parser("java")
tree = parser.parse(bytes(source_code, "utf8"))
root_node = tree.root_node
if root_node.has_error:
raise Exception("Error parsing java the source code.")
def get_text(node):
"""Returns the text represented by the node."""
return source_code[node.start_byte : node.end_byte]
def traverse_node(node, nested=False):
if node.type == "string_literal":
if nested:
return get_text(node)
# Strip surrounding quotes from string literals
return get_text(node)[1:-1]
elif node.type == "character_literal":
if nested:
return get_text(node)
# Strip surrounding single quotes from character literals
return get_text(node)[1:-1]
"""Traverse the node to collect texts for complex structures."""
if node.type in [
"identifier",
"class_literal",
"type_identifier",
"method_invocation",
]:
return get_text(node)
elif node.type == "array_creation_expression":
# Handle array creation expression specifically
type_node = node.child_by_field_name("type")
value_node = node.child_by_field_name("value")
type_text = traverse_node(type_node, True)
value_text = traverse_node(value_node, True)
return f"new {type_text}[]{value_text}"
elif node.type == "object_creation_expression":
# Handle object creation expression specifically
type_node = node.child_by_field_name("type")
arguments_node = node.child_by_field_name("arguments")
type_text = traverse_node(type_node, True)
if arguments_node:
# Process each argument carefully, avoiding unnecessary punctuation
argument_texts = []
for child in arguments_node.children:
if child.type not in [
",",
"(",
")",
]: # Exclude commas and parentheses
argument_text = traverse_node(child, True)
argument_texts.append(argument_text)
arguments_text = ", ".join(argument_texts)
return f"new {type_text}({arguments_text})"
else:
return f"new {type_text}()"
elif node.type == "set":
# Handling sets specifically
items = [traverse_node(n, True) for n in node.children if n.type not in [",", "set"]]
return "{" + ", ".join(items) + "}"
elif node.child_count > 0:
return "".join(traverse_node(child, True) for child in node.children)
else:
return get_text(node)
def extract_arguments(args_node):
arguments = {}
for child in args_node.children:
if child.type == "assignment_expression":
# For named parameters
name_node, value_node = child.children[0], child.children[2]
name = get_text(name_node)
value = traverse_node(value_node)
if name in arguments:
if not isinstance(arguments[name], list):
arguments[name] = [arguments[name]]
arguments[name].append(value)
else:
arguments[name] = value
# arguments.append({'name': name, 'value': value})
elif child.type in ["identifier", "class_literal", "set"]:
# For unnamed parameters and handling sets
value = traverse_node(child)
if None in arguments:
if not isinstance(arguments[None], list):
arguments[None] = [arguments[None]]
arguments[None].append(value)
else:
arguments[None] = value
return arguments
def traverse(node):
if node.type == "method_invocation":
# Extract the function name and its arguments
method_name = get_text(node.child_by_field_name("name"))
class_name_node = node.child_by_field_name("object")
if class_name_node:
class_name = get_text(class_name_node)
function_name = f"{class_name}.{method_name}"
else:
function_name = method_name
arguments_node = node.child_by_field_name("arguments")
if arguments_node:
arguments = extract_arguments(arguments_node)
for key, value in arguments.items():
if isinstance(value, list):
raise Exception("Error: Multiple arguments with the same name are not supported.")
return [{function_name: arguments}]
else:
for child in node.children:
result = traverse(child)
if result:
return result
result = traverse(root_node)
return result if result else {}
def parse_javascript_function_call(source_code):
if not source_code.endswith(";"):
source_code += ";" # Necessary for the parser not to register an error
parser = get_parser("javascript")
# Parse the source code
tree = parser.parse(bytes(source_code, "utf8"))
root_node = tree.root_node
if root_node.has_error:
raise Exception("Error js parsing the source code.")
# Function to recursively extract argument details
def extract_arguments(node):
args = {}
for child in node.children:
if child.type == "assignment_expression":
# Extract left (name) and right (value) parts of the assignment
name = child.children[0].text.decode("utf-8")
value = child.children[2].text.decode("utf-8")
if (value.startswith('"') and value.endswith('"')) or (value.startswith("'") and value.endswith("'")):
value = value[1:-1] # Trim the quotation marks
if name in args:
if not isinstance(args[name], list):
args[name] = [args[name]]
args[name].append(value)
else:
args[name] = value
elif child.type == "identifier" or child.type == "true":
# Handle non-named arguments and boolean values
value = child.text.decode("utf-8")
if None in args:
if not isinstance(args[None], list):
args[None] = [args[None]]
args[None].append(value)
else:
args[None] = value
return args
# Find the function call and extract its name and arguments
if root_node.type == "program":
for child in root_node.children:
if child.type == "expression_statement":
for sub_child in child.children:
if sub_child.type == "call_expression":
function_name = sub_child.children[0].text.decode("utf8")
arguments_node = sub_child.children[1]
parameters = extract_arguments(arguments_node)
for key, value in parameters.items():
if isinstance(value, list):
raise Exception("Error: Multiple arguments with the same name are not supported.")
result = [{function_name: parameters}]
return result
def ast_parse(input_str, language="Python"):
if language == "Python":
cleaned_input = input_str.strip("[]'")
parsed = ast.parse(cleaned_input, mode="eval")
extracted = []
if isinstance(parsed.body, ast.Call):
extracted.append(resolve_ast_call(parsed.body))
else:
for elem in parsed.body.elts:
extracted.append(resolve_ast_call(elem))
return extracted
elif language == "Java":
return parse_java_function_call(input_str[1:-1]) # Remove the [ and ] from the string
elif language == "JavaScript":
return parse_javascript_function_call(input_str[1:-1])
else:
raise NotImplementedError(f"Unsupported language: {language}")
def resolve_ast_call(elem):
# Handle nested attributes for deeply nested module paths
func_parts = []
func_part = elem.func
while isinstance(func_part, ast.Attribute):
func_parts.append(func_part.attr)
func_part = func_part.value
if isinstance(func_part, ast.Name):
func_parts.append(func_part.id)
func_name = ".".join(reversed(func_parts))
args_dict = {}
# Parse when args are simply passed as an unnamed dictionary arg
for arg in elem.args:
if isinstance(arg, ast.Dict):
for key, value in zip(arg.keys, arg.values):
if isinstance(key, ast.Constant):
arg_name = key.value
output = resolve_ast_by_type(value)
args_dict[arg_name] = output
for arg in elem.keywords:
output = resolve_ast_by_type(arg.value)
args_dict[arg.arg] = output
return {func_name: args_dict}
def resolve_ast_by_type(value):
if isinstance(value, ast.Constant):
if value.value is Ellipsis:
output = "..."
else:
output = value.value
elif isinstance(value, ast.UnaryOp):
output = -value.operand.value
elif isinstance(value, ast.List):
output = [resolve_ast_by_type(v) for v in value.elts]
elif isinstance(value, ast.Dict):
output = {resolve_ast_by_type(k): resolve_ast_by_type(v) for k, v in zip(value.keys, value.values)}
elif isinstance(value, ast.NameConstant): # Added this condition to handle boolean values
output = value.value
elif isinstance(value, ast.BinOp): # Added this condition to handle function calls as arguments
output = eval(ast.unparse(value))
elif isinstance(value, ast.Name):
output = value.id
elif isinstance(value, ast.Call):
if len(value.keywords) == 0:
output = ast.unparse(value)
else:
output = resolve_ast_call(value)
elif isinstance(value, ast.Tuple):
output = tuple(resolve_ast_by_type(v) for v in value.elts)
elif isinstance(value, ast.Lambda):
output = eval(ast.unparse(value.body[0].value))
elif isinstance(value, ast.Ellipsis):
output = "..."
elif isinstance(value, ast.Subscript):
try:
output = ast.unparse(value.body[0].value)
except:
output = ast.unparse(value.value) + "[" + ast.unparse(value.slice) + "]"
else:
raise Exception(f"Unsupported AST type: {type(value)}")
return output
def decode_ast(result, language="Python"):
func = result
func = func.replace("\n", "") # remove new line characters
if not func.startswith("["):
func = "[" + func
if not func.endswith("]"):
func = func + "]"
decoded_output = ast_parse(func, language)
return decoded_output
def decode_execute(result):
func = result
func = func.replace("\n", "") # remove new line characters
if not func.startswith("["):
func = "[" + func
if not func.endswith("]"):
func = func + "]"
decode_output = ast_parse(func)
execution_list = []
for function_call in decode_output:
for key, value in function_call.items():
execution_list.append(f"{key}({','.join([f'{k}={repr(v)}' for k, v in value.items()])})")
return execution_list

989
llama_stack/providers/inline/scoring/basic/utils/bfcl/checker.py
View file

@ -1,989 +0,0 @@
# ruff: noqa
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import json
import re
import time
from typing import Any
# Comment out for now until we actually use the rest checker in evals
# import requests # Do not remove this import even though it seems to be unused. It's used in the executable_checker_rest function.
class NoAPIKeyError(Exception):
def __init__(self):
self.message = "❗️Please fill in the API keys in the function_credential_config.json file. If you do not provide the API keys, the executable test category results will be inaccurate."
super().__init__(self.message)
REAL_TIME_MATCH_ALLOWED_DIFFERENCE = 0.2
JAVA_TYPE_CONVERSION = {
"byte": int,
"short": int,
"integer": int,
"float": float,
"double": float,
"long": int,
"boolean": bool,
"char": str,
"Array": list,
"ArrayList": list,
"Set": set,
"HashMap": dict,
"Hashtable": dict,
"Queue": list, # this can be `queue.Queue` as well, for simplicity we check with list
"Stack": list,
"String": str,
"any": str,
}
JS_TYPE_CONVERSION = {
"String": str,
"integer": int,
"float": float,
"Bigint": int,
"Boolean": bool,
"dict": dict,
"array": list,
"any": str,
}
# We switch to conditional import for the following two imports to avoid unnecessary installations.
# User doesn't need to setup the tree-sitter packages if they are not running the test for that language.
# from js_type_converter import js_type_converter
# from java_type_converter import java_type_converter
PYTHON_TYPE_MAPPING = {
"string": str,
"integer": int,
"float": float,
"boolean": bool,
"array": list,
"tuple": list,
"dict": dict,
"any": str,
}
# This is the list of types that we need to recursively check its values
PYTHON_NESTED_TYPE_CHECK_LIST = ["array", "tuple"]
NESTED_CONVERSION_TYPE_LIST = ["Array", "ArrayList", "array"]
#### Helper functions for AST ####
def find_description(func_descriptions, name):
if type(func_descriptions) == list:
for func_description in func_descriptions:
if func_description["name"] == name:
return func_description
return None
else:
# it is a dict, there is only one function
return func_descriptions
def get_possible_answer_type(possible_answer: list):
for answer in possible_answer:
if answer != "": # Optional parameter
return type(answer)
return None
def type_checker(
param: str,
value,
possible_answer: list,
expected_type_description: str,
expected_type_converted,
nested_type_converted,
):
# NOTE: This type checker only supports nested type checking for one level deep.
# We didn't implement recursive type checking for nested types, as it's not needed for the current use case and it's very complex.
result: Any = {
"valid": True,
"error": [],
"is_variable": False,
"error_type": "type_error:simple",
}
is_variable = False
# check for the case where a variable is used instead of a actual value.
# use the type in possible_answer as the expected type
possible_answer_type = get_possible_answer_type(possible_answer)
# if possible_answer only contains optional parameters, we can't determine the type
if possible_answer_type != None:
# we are being precise here.
# in fact, possible_answer_type should always be string, as that's how we treat varibale in possible_answer
if possible_answer_type != expected_type_converted:
is_variable = True
# value is the same type as in function description
if type(value) == expected_type_converted:
# We don't need to do recursive check for simple types
if nested_type_converted == None:
result["is_variable"] = is_variable
return result
else:
for possible_answer_item in possible_answer:
flag = True # Each parameter should match to at least one possible answer type.
# Here, we assume that each item should be the same type. We could also relax it.
if type(possible_answer_item) == list:
for value_item in value:
checker_result = type_checker(
param,
value_item,
possible_answer_item,
str(nested_type_converted),
nested_type_converted,
None,
)
if not checker_result["valid"]:
flag = False
break
if flag:
return {"valid": True, "error": [], "is_variable": is_variable}
result["valid"] = False
result["error"] = [
f"Nested type checking failed for parameter {repr(param)}. Expected outer type {expected_type_description} with inner type {str(nested_type_converted)}. Parameter value: {repr(value)}."
]
result["error_type"] = "type_error:nested"
# value is not as expected, check for the case where a variable is used instead of a actual value
# use the type in possible_answer as the expected type
possible_answer_type = get_possible_answer_type(possible_answer)
# if possible_answer only contains optional parameters, we can't determine the type
if possible_answer_type != None:
# we are being precise here.
# in fact, possible_answer_type should always be string, as that's how we treat varibale in possible_answer
if type(value) == possible_answer_type:
result["is_variable"] = True
return result
result["valid"] = False
result["error"].append(
f"Incorrect type for parameter {repr(param)}. Expected type {expected_type_description}, got {type(value).__name__}. Parameter value: {repr(value)}."
)
result["error_type"] = "type_error:simple"
return result
def standardize_string(input_string: str):
# This function standardizes the string by removing all the spaces, ",./-_*^" punctuation, and converting it to lowercase
# It will also convert all the single quotes to double quotes
# This is used to compare the model output with the possible answers
# We don't want to punish model for answer like April 1, 2024 vs April 1,2024, vs April 1 2024
regex_string = r"[ \,\.\/\-\_\*\^]"
return re.sub(regex_string, "", input_string).lower().replace("'", '"')
def string_checker(param: str, model_output: str, possible_answer: list):
standardize_possible_answer = []
standardize_model_output = standardize_string(model_output)
for i in range(len(possible_answer)):
if type(possible_answer[i]) == str:
standardize_possible_answer.append(standardize_string(possible_answer[i]))
if standardize_model_output not in standardize_possible_answer:
return {
"valid": False,
"error": [
f"Invalid value for parameter {repr(param)}: {repr(model_output)}. Expected one of {possible_answer}. Case insensitive."
],
"error_type": "value_error:string",
}
return {"valid": True, "error": []}
def list_checker(param: str, model_output: list, possible_answer: list):
# Convert the tuple to a list
standardize_model_output = list(model_output)
# If the element in the list is a string, we need to standardize it
for i in range(len(standardize_model_output)):
if type(standardize_model_output[i]) == str:
standardize_model_output[i] = standardize_string(model_output[i])
standardize_possible_answer: Any = []
# We also need to standardize the possible answers
for i in range(len(possible_answer)):
standardize_possible_answer.append([])
for j in range(len(possible_answer[i])):
if type(possible_answer[i][j]) == str:
standardize_possible_answer[i].append(standardize_string(possible_answer[i][j]))
else:
standardize_possible_answer[i].append(possible_answer[i][j])
if standardize_model_output not in standardize_possible_answer:
return {
"valid": False,
"error": [
f"Invalid value for parameter {repr(param)}: {repr(model_output)}. Expected one of {possible_answer}."
],
"error_type": "value_error:list/tuple",
}
return {"valid": True, "error": []}
def dict_checker(param: str, model_output: dict, possible_answers: list):
# This function works for simple dictionaries, but not dictionaries with nested dictionaries.
# The current dataset only contains simple dictionaries, so this is sufficient.
result = {"valid": False, "error": [], "error_type": "dict_checker:unclear"}
for i in range(len(possible_answers)):
if possible_answers[i] == "":
continue
result = {"valid": False, "error": [], "error_type": "dict_checker:unclear"}
flag = True
possible_answer = possible_answers[i]
# possible_anwer is a single dictionary
for key, value in model_output.items():
if key not in possible_answer:
result["valid"] = False
result["error"].append(f"Unexpected dict key parameter: '{key}'.") # type: ignore[attr-defined]
result["error_type"] = "value_error:dict_key"
flag = False
break
standardize_value = value
# If the value is a string, we need to standardize it
if type(value) == str:
standardize_value = standardize_string(value)
# We also need to standardize the possible answers if they are string
standardize_possible_answer = []
for i in range(len(possible_answer[key])):
if type(possible_answer[key][i]) == str:
standardize_possible_answer.append(standardize_string(possible_answer[key][i]))
else:
standardize_possible_answer.append(possible_answer[key][i])
if standardize_value not in standardize_possible_answer:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Invalid value for parameter {repr(key)}: {repr(value)}. Expected one of {standardize_possible_answer}."
)
result["error_type"] = "value_error:dict_value"
flag = False
break
for key, value in possible_answer.items():
if key not in model_output and "" not in value:
result["valid"] = False
result["error"].append(f"Missing dict key parameter: '{key}'.") # type: ignore[attr-defined]
result["error_type"] = "value_error:dict_key"
flag = False
break
if flag:
return {"valid": True, "error": []}
return result
def list_dict_checker(param: str, model_output: list, possible_answers: list):
# This function takes in a list of dictionaries and checks if each dictionary is valid
# The order of the dictionaries in the list must match the order of the possible answers
result = {"valid": False, "error": [], "error_type": "list_dict_checker:unclear"}
for answer_index in range(len(possible_answers)):
flag = True # True means so far, all dictionaries are valid
# Only proceed if the number of dictionaries in the list matches the number of dictionaries in the possible answers
if len(model_output) != len(possible_answers[answer_index]):
result["valid"] = False
result["error"] = ["Wrong number of dictionaries in the list."]
result["error_type"] = "value_error:list_dict_count"
flag = False
continue
for dict_index in range(len(model_output)):
result = dict_checker(
param,
model_output[dict_index],
[possible_answers[answer_index][dict_index]],
)
if not result["valid"]:
flag = False
break
if flag:
return {"valid": True, "error": []}
return result
def simple_function_checker(
func_description: dict,
model_output: dict,
possible_answer: dict,
language: str,
model_name: str,
):
possible_answer = list(possible_answer.values())[0]
# Extract function name and parameters details
func_name = func_description["name"]
param_details = func_description["parameters"]["properties"]
required_params = func_description["parameters"]["required"]
# Initialize a result dictionary
result = {
"valid": True,
"error": [],
"error_type": "simple_function_checker:unclear",
}
# Check if function name matches
if func_name not in model_output:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Function name {repr(func_name)} not found in model output."
)
result["error_type"] = "simple_function_checker:wrong_func_name"
return result
model_params = model_output[func_name]
# Check for required parameters in model output
for param in required_params:
if param not in model_params:
result["valid"] = False
result["error"].append(f"Missing required parameter: {repr(param)}.") # type: ignore[attr-defined]
result["error_type"] = "simple_function_checker:missing_required"
return result
# Validate types and values for each parameter in model output
for param, value in model_params.items():
if param not in param_details or param not in possible_answer:
result["valid"] = False
result["error"].append(f"Unexpected parameter: {repr(param)}.") # type: ignore[attr-defined]
result["error_type"] = "simple_function_checker:unexpected_param"
return result
full_param_details = param_details[param]
expected_type_description = full_param_details["type"] # This is a string
is_variable = False
nested_type_converted = None
if language == "Java":
from evals.utils.bfcl.java_type_converter import java_type_converter
expected_type_converted = JAVA_TYPE_CONVERSION[expected_type_description]
if expected_type_description in JAVA_TYPE_CONVERSION:
if type(value) != str:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Incorrect type for parameter {repr(param)}. Expected type String, got {type(value).__name__}. Parameter value: {repr(value)}."
)
result["error_type"] = "type_error:java"
return result
if expected_type_description in NESTED_CONVERSION_TYPE_LIST:
nested_type = param_details[param]["items"]["type"]
nested_type_converted = JAVA_TYPE_CONVERSION[nested_type]
value = java_type_converter(value, expected_type_description, nested_type)
else:
value = java_type_converter(value, expected_type_description)
elif language == "JavaScript":
from evals.utils.bfcl.js_type_converter import js_type_converter
expected_type_converted = JS_TYPE_CONVERSION[expected_type_description]
if expected_type_description in JS_TYPE_CONVERSION:
if type(value) != str:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Incorrect type for parameter {repr(param)}. Expected type String, got {type(value).__name__}. Parameter value: {repr(value)}."
)
result["error_type"] = "type_error:js"
return result
if expected_type_description in NESTED_CONVERSION_TYPE_LIST:
nested_type = param_details[param]["items"]["type"]
nested_type_converted = JS_TYPE_CONVERSION[nested_type]
value = js_type_converter(value, expected_type_description, nested_type)
else:
value = js_type_converter(value, expected_type_description)
elif language == "Python":
expected_type_converted = PYTHON_TYPE_MAPPING[expected_type_description]
if expected_type_description in PYTHON_NESTED_TYPE_CHECK_LIST:
nested_type = param_details[param]["items"]["type"]
nested_type_converted = PYTHON_TYPE_MAPPING[nested_type]
# We convert all tuple value to list when the expected type is tuple.
# The conversion is necessary because any tuple in the possible answer would become a list after being processed through json.dump() and json.load().
# This does introduce some false positive (eg, when the model provides a list value instead of tuple). We hope to find a better solution in the future.
if expected_type_description == "tuple" and type(value) == tuple:
value = list(value)
# Allow python auto conversion from int to float
if language == "Python" and expected_type_description == "float" and type(value) == int:
value = float(value)
# Type checking
# In fact, we only check for Python here.
# Type check for other languages are handled by the type converter, and so their value (after conversion) is always correct.
type_check_result = type_checker(
param,
value,
possible_answer[param],
expected_type_description,
expected_type_converted,
nested_type_converted,
)
is_variable = type_check_result["is_variable"]
if not type_check_result["valid"]:
return type_check_result
# It doesn't make sense to special handle dictionaries and list of dictionaries if the value is a variable.
# We can just treat the variable as a string and use the normal flow.
if not is_variable:
# Special handle for dictionaries
if expected_type_converted == dict:
result = dict_checker(param, value, possible_answer[param])
if not result["valid"]:
return result
continue
# Special handle for list of dictionaries
elif expected_type_converted == list and nested_type_converted == dict:
result = list_dict_checker(param, value, possible_answer[param])
if not result["valid"]:
return result
continue
# Special handle for strings
elif expected_type_converted == str:
# We don't check for case sensitivity for string, as long as it's not a variable
result = string_checker(param, value, possible_answer[param])
if not result["valid"]:
return result
continue
elif expected_type_converted == list:
result = list_checker(param, value, possible_answer[param])
if not result["valid"]:
return result
continue
# Check if the value is within the possible answers
if value not in possible_answer[param]:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Invalid value for parameter {repr(param)}: {repr(value)}. Expected one of {possible_answer[param]}."
)
result["error_type"] = "value_error:others"
return result
# Check for optional parameters not provided but allowed
for param in possible_answer:
if param not in model_params and "" not in possible_answer[param]:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Optional parameter {repr(param)} not provided and not marked as optional."
)
result["error_type"] = "simple_function_checker:missing_optional"
return result
return result
def parallel_function_checker_enforce_order(
func_descriptions: list,
model_output: list,
possible_answers: dict,
language: str,
model_name: str,
):
if len(model_output) != len(possible_answers):
return {
"valid": False,
"error": ["Wrong number of functions."],
"error_type": "parallel_function_checker_enforce_order:wrong_count",
}
func_name_list = list(possible_answers.keys())
possible_answers_list = []
for key, value in possible_answers.items():
possible_answers_list.append({key: value})
for i in range(len(possible_answers_list)):
func_description = find_description(func_descriptions, func_name_list[i])
result = simple_function_checker(
func_description,
model_output[i],
possible_answers_list[i],
language,
model_name,
)
if not result["valid"]:
return result
return {"valid": True, "error": []}
def parallel_function_checker_no_order(
func_descriptions: list,
model_output: list,
possible_answers: list,
language: str,
model_name: str,
):
if len(model_output) != len(possible_answers):
return {
"valid": False,
"error": ["Wrong number of functions."],
"error_type": "parallel_function_checker_no_order:wrong_count",
}
matched_indices = []
# We go throught the possible answers one by one, and eliminate the model output that matches the possible answer
# It must be this way because we need ground truth to fetch the correct function description
for i in range(len(possible_answers)):
# possible_answers[i] is a dictionary with only one key
func_name_expected = list(possible_answers[i].keys())[0]
func_description = find_description(func_descriptions, func_name_expected)
all_errors = []
for index in range(len(model_output)):
if index in matched_indices:
continue
result = simple_function_checker(
func_description,
model_output[index],
possible_answers[i],
language,
model_name,
)
if result["valid"]:
matched_indices.append(index)
break
else:
all_errors.append(
{
f"Model Result Index {index}": {
"sub_error": result["error"],
"sub_error_type": result["error_type"],
"model_output_item": model_output[index],
"possible_answer_item": possible_answers[i],
}
}
)
if not result["valid"]:
considered_indices = [i for i in range(len(model_output)) if i not in matched_indices]
all_errors.insert(
0,
f"Could not find a matching function among index {considered_indices} of model output for index {i} of possible answers.", # type: ignore[arg-type]
)
return {
"valid": False,
"error": all_errors,
"error_type": "parallel_function_checker_no_order:cannot_find_match",
}
return {"valid": True, "error": []}
def multiple_function_checker(
func_descriptions: list,
model_output: list,
possible_answers: list,
language: str,
model_name: str,
):
if len(model_output) != len(possible_answers):
return {
"valid": False,
"error": ["Wrong number of functions."],
"error_type": "multiple_function_checker:wrong_count",
}
# possible_answers is a list of only one dictionary with only one key
func_name_expected = list(possible_answers[0].keys())[0]
func_description = find_description(func_descriptions, func_name_expected)
return simple_function_checker(
func_description,
model_output[0],
possible_answers[0],
language,
model_name,
)
def patten_matcher(exec_output, expected_result, function_call, is_sanity_check):
result = {"valid": True, "error": [], "error_type": "executable_checker:unclear"}
if type(exec_output) != type(expected_result):
return {
"valid": False,
"error": [
f"Wrong execution result type for {repr(function_call)}. Expected type: {type(expected_result)}, but got: {type(exec_output)}."
],
"error_type": "executable_checker:wrong_result_type",
"model_executed_output": exec_output,
}
if type(exec_output) == dict:
# We loose the requirement for the sanity check as the expected result used in the sanity check might not be the most up-to-date one.
# This happens when the key is a timestamp or a random number.
if is_sanity_check:
if len(exec_output) != len(expected_result):
return {
"valid": False,
"error": [
f"Wrong execution result pattern for {repr(function_call)}. Expect type Dict, but wrong number of elements in the output. Expected length: {len(expected_result)}, but got: {len(exec_output)}."
],
"error_type": "executable_checker:wrong_result_type:dict_length",
"model_executed_output": exec_output,
}
else:
return result
for key, value in expected_result.items():
if key not in exec_output:
return {
"valid": False,
"error": [
f"Wrong execution result pattern for {repr(function_call)}. Expect type Dict, but key {repr(key)} not found in the model output."
],
"error_type": "executable_checker:wrong_result_type:dict_key_not_found",
"model_executed_output": exec_output,
}
for key, value in exec_output.items():
if key not in expected_result:
return {
"valid": False,
"error": [
f"Wrong execution result pattern for {repr(function_call)}. Expect type Dict, but key {repr(key)} not expected in the model output."
],
"error_type": "executable_checker:wrong_result_type:dict_extra_key",
"model_executed_output": exec_output,
}
if type(exec_output) == list:
if len(exec_output) != len(expected_result):
return {
"valid": False,
"error": [
f"Wrong execution result pattern for {repr(function_call)}. Expect type list, but wrong number of elements in the output. Expected length: {len(expected_result)}, but got: {len(exec_output)}."
],
"error_type": "executable_checker:wrong_result_type:list_length",
"model_executed_output": exec_output,
}
return result
#### Helper functions for Exec ####
def executable_checker_simple(
function_call: str,
expected_result,
expected_result_type: str,
is_sanity_check=False,
):
result = {"valid": True, "error": [], "error_type": "executable_checker:unclear"}
exec_dict: Any = {}
try:
exec(
"from executable_python_function import *" + "\nresult=" + function_call,
exec_dict,
)
exec_output = exec_dict["result"]
except NoAPIKeyError as e:
raise e
except Exception as e:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Error in execution: {repr(function_call)}. Error: {str(e)}"
)
result["error_type"] = "executable_checker:execution_error"
return result
# We need to special handle the case where the execution result is a tuple and convert it to a list
# Because when json is stored, the tuple is converted to a list, and so the expected result is a list when loaded from json
if isinstance(exec_output, tuple):
exec_output = list(exec_output)
if expected_result_type == "exact_match":
if exec_output != expected_result:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Wrong execution result for {repr(function_call)}. Expected: {expected_result}, but got: {exec_output}."
)
result["error_type"] = "executable_checker:wrong_result"
result["model_executed_output"] = exec_output
return result
elif expected_result_type == "real_time_match":
# Allow for 5% difference
if (type(expected_result) == float or type(expected_result) == int) and (
type(exec_output) == float or type(exec_output) == int
):
if not (
expected_result * (1 - REAL_TIME_MATCH_ALLOWED_DIFFERENCE)
<= exec_output
<= expected_result * (1 + REAL_TIME_MATCH_ALLOWED_DIFFERENCE)
):
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Wrong execution result for {repr(function_call)}. Expected: {expected_result}, but got: {exec_output}. {REAL_TIME_MATCH_ALLOWED_DIFFERENCE * 100}% difference allowed."
)
result["error_type"] = "executable_checker:wrong_result_real_time"
result["model_executed_output"] = exec_output
return result
else:
result["valid"] = False
result["error"].append( # type: ignore[attr-defined]
f"Wrong execution result for {repr(function_call)}. Expected: {expected_result}, but got: {exec_output}. Type needs to be float or int for real time match criteria."
)
result["error_type"] = "executable_checker:wrong_result_real_time"
result["model_executed_output"] = exec_output
return result
else:
# structural match
pattern_match_result = patten_matcher(exec_output, expected_result, function_call, is_sanity_check)
if not pattern_match_result["valid"]:
return pattern_match_result
return result
def executable_checker_parallel_no_order(
decoded_result: list, expected_exec_result: list, expected_exec_result_type: list
):
if len(decoded_result) != len(expected_exec_result):
return {
"valid": False,
"error": [
f"Wrong number of functions provided. Expected {len(expected_exec_result)}, but got {len(decoded_result)}."
],
"error_type": "value_error:exec_result_count",
}
matched_indices = []
for i in range(len(expected_exec_result)):
all_errors = []
for index in range(len(decoded_result)):
if index in matched_indices:
continue
result = executable_checker_simple(
decoded_result[index],
expected_exec_result[i],
expected_exec_result_type[i],
False,
)
if result["valid"]:
matched_indices.append(index)
break
else:
all_errors.append(
{
f"Model Result Index {index}": {
"sub_error": result["error"],
"sub_error_type": result["error_type"],
"model_executed_output": (
result["model_executed_output"] if "model_executed_output" in result else None
),
}
}
)
if not result["valid"]:
considered_indices = [i for i in range(len(decoded_result)) if i not in matched_indices]
all_errors.insert(
0,
f"Could not find a matching function among index {considered_indices} of model output for index {i} of possible answers.", # type: ignore[arg-type]
)
return {
"valid": False,
"error": all_errors,
"error_type": "executable_checker:cannot_find_match",
}
return {"valid": True, "error": [], "error_type": "executable_checker:unclear"}
#### Main function ####
def executable_checker_rest(func_call, idx):
# Move this here for now to avoid needing to read this file / fix paths to be relative to dataset_dir. Fix when it's actually needed / used.
EVAL_GROUND_TRUTH_PATH = "/mnt/wsfuse/fair_llm_v2/datasets/eval/bfcl/rest-eval-response_v5.jsonl" # Ground truth file for v5 for rest execution
with open(EVAL_GROUND_TRUTH_PATH, "r") as f:
EVAL_GROUND_TRUTH = f.readlines()
if "https://geocode.maps.co" in func_call:
time.sleep(2)
if "requests_get" in func_call:
func_call = func_call.replace("requests_get", "requests.get")
try:
response = eval(func_call)
except Exception as e:
return {
"valid": False,
"error": [f"Execution failed. {str(e)}"],
"error_type": "executable_checker_rest:execution_error",
}
try:
if response.status_code == 200:
eval_GT_json = json.loads(EVAL_GROUND_TRUTH[idx])
try:
if isinstance(eval_GT_json, dict):
if isinstance(response.json(), dict):
if set(eval_GT_json.keys()) == set(response.json().keys()):
return {"valid": True, "error": [], "error_type": ""}
return {
"valid": False,
"error": ["Key inconsistency"],
"error_type": "executable_checker_rest:wrong_key",
}
return {
"valid": False,
"error": [f"Expected dictionary, but got {type(response.json())}"],
"error_type": "executable_checker_rest:wrong_type",
}
elif isinstance(eval_GT_json, list):
if isinstance(response.json(), list):
if len(eval_GT_json) != len(response.json()):
return {
"valid": False,
"error": [f"Response list length inconsistency."],
"error_type": "value_error:exec_result_rest_count",
}
else:
for i in range(len(eval_GT_json)):
if set(eval_GT_json[i].keys()) != set(response.json()[i].keys()):
return {
"valid": False,
"error": [f"Key inconsistency"],
"error_type": "executable_checker_rest:wrong_key",
}
return {"valid": True, "error": []}
else:
return {
"valid": False,
"error": [f"Expected list, but got {type(response.json())}"],
"error_type": "executable_checker_rest:wrong_type",
}
return {
"valid": False,
"error": [f"Expected dict or list, but got {type(response.json())}"],
"error_type": "executable_checker_rest:wrong_type",
}
except Exception as e:
return {
"valid": False,
"error": [
f"Error in execution and type checking. Status code: {response.status_code}. Error: {str(e)}"
],
"error_type": "executable_checker_rest:response_format_error",
}
else:
return {
"valid": False,
"error": [f"Execution result status code is not 200, got {response.status_code}"],
"error_type": "executable_checker_rest:wrong_status_code",
}
except Exception as e:
return {
"valid": False,
"error": [f"Cannot get status code of the response. Error: {str(e)}"],
"error_type": "executable_checker_rest:cannot_get_status_code",
}
def ast_checker(func_description, model_output, possible_answer, language, test_category, model_name):
if "parallel" in test_category:
return parallel_function_checker_no_order(func_description, model_output, possible_answer, language, model_name)
elif "multiple" in test_category:
return multiple_function_checker(func_description, model_output, possible_answer, language, model_name)
else:
if len(model_output) != 1:
return {
"valid": False,
"error": ["Wrong number of functions."],
"error_type": "simple_function_checker:wrong_count",
}
return simple_function_checker(
func_description[0],
model_output[0],
possible_answer[0],
language,
model_name,
)
def exec_checker(decoded_result: list, func_description: dict, test_category: str):
if "multiple" in test_category or "parallel" in test_category:
return executable_checker_parallel_no_order(
decoded_result,
func_description["execution_result"],
func_description["execution_result_type"],
)
else:
if len(decoded_result) != 1:
return {
"valid": False,
"error": ["Wrong number of functions."],
"error_type": "simple_exec_checker:wrong_count",
}
return executable_checker_simple(
decoded_result[0],
func_description["execution_result"][0],
func_description["execution_result_type"][0],
False,
)
def is_empty_output(decoded_output):
# This function is a patch to the ast decoder for relevance detection
# Sometimes the ast decoder will parse successfully, but the input doens't really have a function call
# [], [{}], and anything that is not in function calling format is considered empty (and thus should be marked as correct)
if not is_function_calling_format_output(decoded_output):
return True
if len(decoded_output) == 0:
return True
if len(decoded_output) == 1 and len(decoded_output[0]) == 0:
return True
def is_function_calling_format_output(decoded_output):
# Ensure the output is a list of dictionaries
if type(decoded_output) == list:
for item in decoded_output:
if type(item) != dict:
return False
return True
return False

40
llama_stack/providers/inline/scoring/basic/utils/bfcl/tree_sitter.py
View file

@ -1,40 +0,0 @@
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
"""
Tree-sitter changes its API with unfortunate frequency. Modules that need it should
import it from here so that we can centrally manage things as necessary.
"""
# These currently work with tree-sitter 0.23.0
# NOTE: Don't import tree-sitter or any of the language modules in the main module
# because not all environments have them. Import lazily inside functions where needed.
import importlib
import typing
if typing.TYPE_CHECKING:
import tree_sitter
def get_language(language: str) -> "tree_sitter.Language":
import tree_sitter
language_module_name = f"tree_sitter_{language}"
try:
language_module = importlib.import_module(language_module_name)
except ModuleNotFoundError as exc:
raise ValueError(
f"Language {language} is not found. Please install the tree-sitter-{language} package."
) from exc
return tree_sitter.Language(language_module.language())
def get_parser(language: str, **kwargs) -> "tree_sitter.Parser":
import tree_sitter
lang = get_language(language)
return tree_sitter.Parser(lang, **kwargs)

67
llama_stack/providers/inline/vector_io/sqlite_vec/sqlite_vec.py
View file

@ -30,11 +30,11 @@ from llama_stack.providers.utils.kvstore.api import KVStore
from llama_stack.providers.utils.memory.openai_vector_store_mixin import OpenAIVectorStoreMixin
from llama_stack.providers.utils.memory.vector_store import (
RERANKER_TYPE_RRF,
RERANKER_TYPE_WEIGHTED,
ChunkForDeletion,
EmbeddingIndex,
VectorDBWithIndex,
)
from llama_stack.providers.utils.vector_io.vector_utils import WeightedInMemoryAggregator
logger = get_logger(name=__name__, category="vector_io")
@ -66,59 +66,6 @@ def _create_sqlite_connection(db_path):
return connection
def _normalize_scores(scores: dict[str, float]) -> dict[str, float]:
"""Normalize scores to [0,1] range using min-max normalization."""
if not scores:
return {}
min_score = min(scores.values())
max_score = max(scores.values())
score_range = max_score - min_score
if score_range > 0:
return {doc_id: (score - min_score) / score_range for doc_id, score in scores.items()}
return dict.fromkeys(scores, 1.0)
def _weighted_rerank(
vector_scores: dict[str, float],
keyword_scores: dict[str, float],
alpha: float = 0.5,
) -> dict[str, float]:
"""ReRanker that uses weighted average of scores."""
all_ids = set(vector_scores.keys()) | set(keyword_scores.keys())
normalized_vector_scores = _normalize_scores(vector_scores)
normalized_keyword_scores = _normalize_scores(keyword_scores)
return {
doc_id: (alpha * normalized_keyword_scores.get(doc_id, 0.0))
+ ((1 - alpha) * normalized_vector_scores.get(doc_id, 0.0))
for doc_id in all_ids
}
def _rrf_rerank(
vector_scores: dict[str, float],
keyword_scores: dict[str, float],
impact_factor: float = 60.0,
) -> dict[str, float]:
"""ReRanker that uses Reciprocal Rank Fusion."""
# Convert scores to ranks
vector_ranks = {
doc_id: i + 1 for i, (doc_id, _) in enumerate(sorted(vector_scores.items(), key=lambda x: x[1], reverse=True))
}
keyword_ranks = {
doc_id: i + 1 for i, (doc_id, _) in enumerate(sorted(keyword_scores.items(), key=lambda x: x[1], reverse=True))
}
all_ids = set(vector_scores.keys()) | set(keyword_scores.keys())
rrf_scores = {}
for doc_id in all_ids:
vector_rank = vector_ranks.get(doc_id, float("inf"))
keyword_rank = keyword_ranks.get(doc_id, float("inf"))
# RRF formula: score = 1/(k + r) where k is impact_factor and r is the rank
rrf_scores[doc_id] = (1.0 / (impact_factor + vector_rank)) + (1.0 / (impact_factor + keyword_rank))
return rrf_scores
def _make_sql_identifier(name: str) -> str:
return re.sub(r"[^a-zA-Z0-9_]", "_", name)
@ -398,14 +345,10 @@ class SQLiteVecIndex(EmbeddingIndex):
for chunk, score in zip(keyword_response.chunks, keyword_response.scores, strict=False)
}
# Combine scores using the specified reranker
if reranker_type == RERANKER_TYPE_WEIGHTED:
alpha = reranker_params.get("alpha", 0.5)
combined_scores = _weighted_rerank(vector_scores, keyword_scores, alpha)
else:
# Default to RRF for None, RRF, or any unknown types
impact_factor = reranker_params.get("impact_factor", 60.0)
combined_scores = _rrf_rerank(vector_scores, keyword_scores, impact_factor)
# Combine scores using the reranking utility
combined_scores = WeightedInMemoryAggregator.combine_search_results(
vector_scores, keyword_scores, reranker_type, reranker_params
)
# Sort by combined score and get top k results
sorted_items = sorted(combined_scores.items(), key=lambda x: x[1], reverse=True)

2
llama_stack/providers/registry/inference.py
View file

@ -292,7 +292,7 @@ Available Models:
api=Api.inference,
adapter=AdapterSpec(
adapter_type="watsonx",
pip_packages=["ibm_watson_machine_learning"],
pip_packages=["ibm_watsonx_ai"],
module="llama_stack.providers.remote.inference.watsonx",
config_class="llama_stack.providers.remote.inference.watsonx.WatsonXConfig",
provider_data_validator="llama_stack.providers.remote.inference.watsonx.WatsonXProviderDataValidator",

74
llama_stack/providers/registry/vector_io.py
View file

@ -404,6 +404,60 @@ That means you'll get fast and efficient vector retrieval.
- Easy to use
- Fully integrated with Llama Stack
There are three implementations of search for PGVectoIndex available:
1. Vector Search:
- How it works:
- Uses PostgreSQL's vector extension (pgvector) to perform similarity search
- Compares query embeddings against stored embeddings using Cosine distance or other distance metrics
- Eg. SQL query: SELECT document, embedding <=> %s::vector AS distance FROM table ORDER BY distance
-Characteristics:
- Semantic understanding - finds documents similar in meaning even if they don't share keywords
- Works with high-dimensional vector embeddings (typically 768, 1024, or higher dimensions)
- Best for: Finding conceptually related content, handling synonyms, cross-language search
2. Keyword Search
- How it works:
- Uses PostgreSQL's full-text search capabilities with tsvector and ts_rank
- Converts text to searchable tokens using to_tsvector('english', text). Default language is English.
- Eg. SQL query: SELECT document, ts_rank(tokenized_content, plainto_tsquery('english', %s)) AS score
- Characteristics:
- Lexical matching - finds exact keyword matches and variations
- Uses GIN (Generalized Inverted Index) for fast text search performance
- Scoring: Uses PostgreSQL's ts_rank function for relevance scoring
- Best for: Exact term matching, proper names, technical terms, Boolean-style queries
3. Hybrid Search
- How it works:
- Combines both vector and keyword search results
- Runs both searches independently, then merges results using configurable reranking
- Two reranking strategies available:
- Reciprocal Rank Fusion (RRF) - (default: 60.0)
- Weighted Average - (default: 0.5)
- Characteristics:
- Best of both worlds: semantic understanding + exact matching
- Documents appearing in both searches get boosted scores
- Configurable balance between semantic and lexical matching
- Best for: General-purpose search where you want both precision and recall
4. Database Schema
The PGVector implementation stores data optimized for all three search types:
CREATE TABLE vector_store_xxx (
id TEXT PRIMARY KEY,
document JSONB, -- Original document
embedding vector(dimension), -- For vector search
content_text TEXT, -- Raw text content
tokenized_content TSVECTOR -- For keyword search
);
-- Indexes for performance
CREATE INDEX content_gin_idx ON table USING GIN(tokenized_content); -- Keyword search
-- Vector index created automatically by pgvector
## Usage
To use PGVector in your Llama Stack project, follow these steps:
@ -412,6 +466,25 @@ To use PGVector in your Llama Stack project, follow these steps:
2. Configure your Llama Stack project to use pgvector. (e.g. remote::pgvector).
3. Start storing and querying vectors.
## This is an example how you can set up your environment for using PGVector
1. Export env vars:
```bash
export ENABLE_PGVECTOR=true
export PGVECTOR_HOST=localhost
export PGVECTOR_PORT=5432
export PGVECTOR_DB=llamastack
export PGVECTOR_USER=llamastack
export PGVECTOR_PASSWORD=llamastack
```
2. Create DB:
```bash
psql -h localhost -U postgres -c "CREATE ROLE llamastack LOGIN PASSWORD 'llamastack';"
psql -h localhost -U postgres -c "CREATE DATABASE llamastack OWNER llamastack;"
psql -h localhost -U llamastack -d llamastack -c "CREATE EXTENSION IF NOT EXISTS vector;"
```
## Installation
You can install PGVector using docker:
@ -449,6 +522,7 @@ Weaviate supports:
- Metadata filtering
- Multi-modal retrieval
## Usage
To use Weaviate in your Llama Stack project, follow these steps:

7
llama_stack/providers/remote/files/s3/__init__.py
View file

@ -6,15 +6,14 @@
from typing import Any
from llama_stack.core.datatypes import Api
from llama_stack.core.datatypes import AccessRule, Api
from .config import S3FilesImplConfig
async def get_adapter_impl(config: S3FilesImplConfig, deps: dict[Api, Any]):
async def get_adapter_impl(config: S3FilesImplConfig, deps: dict[Api, Any], policy: list[AccessRule] | None = None):
from .files import S3FilesImpl
# TODO: authorization policies and user separation
impl = S3FilesImpl(config)
impl = S3FilesImpl(config, policy or [])
await impl.initialize()
return impl

187
llama_stack/providers/remote/files/s3/files.py
View file

@ -4,9 +4,9 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import time
import uuid
from typing import Annotated
from datetime import UTC, datetime
from typing import Annotated, Any
import boto3
from botocore.exceptions import BotoCoreError, ClientError, NoCredentialsError
@ -15,14 +15,17 @@ from fastapi import File, Form, Response, UploadFile
from llama_stack.apis.common.errors import ResourceNotFoundError
from llama_stack.apis.common.responses import Order
from llama_stack.apis.files import (
ExpiresAfter,
Files,
ListOpenAIFileResponse,
OpenAIFileDeleteResponse,
OpenAIFileObject,
OpenAIFilePurpose,
)
from llama_stack.core.datatypes import AccessRule
from llama_stack.providers.utils.sqlstore.api import ColumnDefinition, ColumnType
from llama_stack.providers.utils.sqlstore.sqlstore import SqlStore, sqlstore_impl
from llama_stack.providers.utils.sqlstore.authorized_sqlstore import AuthorizedSqlStore
from llama_stack.providers.utils.sqlstore.sqlstore import sqlstore_impl
from .config import S3FilesImplConfig
@ -83,22 +86,85 @@ async def _create_bucket_if_not_exists(client: boto3.client, config: S3FilesImpl
raise RuntimeError(f"Failed to access S3 bucket '{config.bucket_name}': {e}") from e
def _make_file_object(
*,
id: str,
filename: str,
purpose: str,
bytes: int,
created_at: int,
expires_at: int,
**kwargs: Any, # here to ignore any additional fields, e.g. extra fields from AuthorizedSqlStore
) -> OpenAIFileObject:
"""
Construct an OpenAIFileObject and normalize expires_at.
If expires_at is greater than the max we treat it as no-expiration and
return None for expires_at.
The OpenAI spec says expires_at type is Integer, but the implementation
will return None for no expiration.
"""
obj = OpenAIFileObject(
id=id,
filename=filename,
purpose=OpenAIFilePurpose(purpose),
bytes=bytes,
created_at=created_at,
expires_at=expires_at,
)
if obj.expires_at is not None and obj.expires_at > (obj.created_at + ExpiresAfter.MAX):
obj.expires_at = None # type: ignore
return obj
class S3FilesImpl(Files):
"""S3-based implementation of the Files API."""
# TODO: implement expiration, for now a silly offset
_SILLY_EXPIRATION_OFFSET = 100 * 365 * 24 * 60 * 60
def __init__(self, config: S3FilesImplConfig) -> None:
def __init__(self, config: S3FilesImplConfig, policy: list[AccessRule]) -> None:
self._config = config
self.policy = policy
self._client: boto3.client | None = None
self._sql_store: SqlStore | None = None
self._sql_store: AuthorizedSqlStore | None = None
def _now(self) -> int:
"""Return current UTC timestamp as int seconds."""
return int(datetime.now(UTC).timestamp())
async def _get_file(self, file_id: str, return_expired: bool = False) -> dict[str, Any]:
where: dict[str, str | dict] = {"id": file_id}
if not return_expired:
where["expires_at"] = {">": self._now()}
if not (row := await self.sql_store.fetch_one("openai_files", policy=self.policy, where=where)):
raise ResourceNotFoundError(file_id, "File", "files.list()")
return row
async def _delete_file(self, file_id: str) -> None:
"""Delete a file from S3 and the database."""
try:
self.client.delete_object(
Bucket=self._config.bucket_name,
Key=file_id,
)
except ClientError as e:
if e.response["Error"]["Code"] != "NoSuchKey":
raise RuntimeError(f"Failed to delete file from S3: {e}") from e
await self.sql_store.delete("openai_files", where={"id": file_id})
async def _delete_if_expired(self, file_id: str) -> None:
"""If the file exists and is expired, delete it."""
if row := await self._get_file(file_id, return_expired=True):
if (expires_at := row.get("expires_at")) and expires_at <= self._now():
await self._delete_file(file_id)
async def initialize(self) -> None:
self._client = _create_s3_client(self._config)
await _create_bucket_if_not_exists(self._client, self._config)
self._sql_store = sqlstore_impl(self._config.metadata_store)
self._sql_store = AuthorizedSqlStore(sqlstore_impl(self._config.metadata_store))
await self._sql_store.create_table(
"openai_files",
{
@ -121,7 +187,7 @@ class S3FilesImpl(Files):
return self._client
@property
def sql_store(self) -> SqlStore:
def sql_store(self) -> AuthorizedSqlStore:
assert self._sql_store is not None, "Provider not initialized"
return self._sql_store
@ -129,27 +195,47 @@ class S3FilesImpl(Files):
self,
file: Annotated[UploadFile, File()],
purpose: Annotated[OpenAIFilePurpose, Form()],
expires_after_anchor: Annotated[str | None, Form(alias="expires_after[anchor]")] = None,
expires_after_seconds: Annotated[int | None, Form(alias="expires_after[seconds]")] = None,
) -> OpenAIFileObject:
file_id = f"file-{uuid.uuid4().hex}"
filename = getattr(file, "filename", None) or "uploaded_file"
created_at = int(time.time())
expires_at = created_at + self._SILLY_EXPIRATION_OFFSET
created_at = self._now()
expires_after = None
if expires_after_anchor is not None or expires_after_seconds is not None:
# we use ExpiresAfter to validate input
expires_after = ExpiresAfter(
anchor=expires_after_anchor, # type: ignore[arg-type]
seconds=expires_after_seconds, # type: ignore[arg-type]
)
# the default is no expiration.
# to implement no expiration we set an expiration beyond the max.
# we'll hide this fact from users when returning the file object.
expires_at = created_at + ExpiresAfter.MAX * 42
# the default for BATCH files is 30 days, which happens to be the expiration max.
if purpose == OpenAIFilePurpose.BATCH:
expires_at = created_at + ExpiresAfter.MAX
if expires_after is not None:
expires_at = created_at + expires_after.seconds
content = await file.read()
file_size = len(content)
await self.sql_store.insert(
"openai_files",
{
entry: dict[str, Any] = {
"id": file_id,
"filename": filename,
"purpose": purpose.value,
"bytes": file_size,
"created_at": created_at,
"expires_at": expires_at,
},
)
}
await self.sql_store.insert("openai_files", entry)
try:
self.client.put_object(
@ -163,14 +249,7 @@ class S3FilesImpl(Files):
raise RuntimeError(f"Failed to upload file to S3: {e}") from e
return OpenAIFileObject(
id=file_id,
filename=filename,
purpose=purpose,
bytes=file_size,
created_at=created_at,
expires_at=expires_at,
)
return _make_file_object(**entry)
async def openai_list_files(
self,
@ -183,29 +262,20 @@ class S3FilesImpl(Files):
if not order:
order = Order.desc
where_conditions = {}
where_conditions: dict[str, Any] = {"expires_at": {">": self._now()}}
if purpose:
where_conditions["purpose"] = purpose.value
paginated_result = await self.sql_store.fetch_all(
table="openai_files",
where=where_conditions if where_conditions else None,
policy=self.policy,
where=where_conditions,
order_by=[("created_at", order.value)],
cursor=("id", after) if after else None,
limit=limit,
)
files = [
OpenAIFileObject(
id=row["id"],
filename=row["filename"],
purpose=OpenAIFilePurpose(row["purpose"]),
bytes=row["bytes"],
created_at=row["created_at"],
expires_at=row["expires_at"],
)
for row in paginated_result.data
]
files = [_make_file_object(**row) for row in paginated_result.data]
return ListOpenAIFileResponse(
data=files,
@ -216,41 +286,20 @@ class S3FilesImpl(Files):
)
async def openai_retrieve_file(self, file_id: str) -> OpenAIFileObject:
row = await self.sql_store.fetch_one("openai_files", where={"id": file_id})
if not row:
raise ResourceNotFoundError(file_id, "File", "files.list()")
return OpenAIFileObject(
id=row["id"],
filename=row["filename"],
purpose=OpenAIFilePurpose(row["purpose"]),
bytes=row["bytes"],
created_at=row["created_at"],
expires_at=row["expires_at"],
)
await self._delete_if_expired(file_id)
row = await self._get_file(file_id)
return _make_file_object(**row)
async def openai_delete_file(self, file_id: str) -> OpenAIFileDeleteResponse:
row = await self.sql_store.fetch_one("openai_files", where={"id": file_id})
if not row:
raise ResourceNotFoundError(file_id, "File", "files.list()")
try:
self.client.delete_object(
Bucket=self._config.bucket_name,
Key=row["id"],
)
except ClientError as e:
if e.response["Error"]["Code"] != "NoSuchKey":
raise RuntimeError(f"Failed to delete file from S3: {e}") from e
await self.sql_store.delete("openai_files", where={"id": file_id})
await self._delete_if_expired(file_id)
_ = await self._get_file(file_id) # raises if not found
await self._delete_file(file_id)
return OpenAIFileDeleteResponse(id=file_id, deleted=True)
async def openai_retrieve_file_content(self, file_id: str) -> Response:
row = await self.sql_store.fetch_one("openai_files", where={"id": file_id})
if not row:
raise ResourceNotFoundError(file_id, "File", "files.list()")
await self._delete_if_expired(file_id)
row = await self._get_file(file_id)
try:
response = self.client.get_object(
@ -261,7 +310,7 @@ class S3FilesImpl(Files):
content = response["Body"].read()
except ClientError as e:
if e.response["Error"]["Code"] == "NoSuchKey":
await self.sql_store.delete("openai_files", where={"id": file_id})
await self._delete_file(file_id)
raise ResourceNotFoundError(file_id, "File", "files.list()") from e
raise RuntimeError(f"Failed to download file from S3: {e}") from e

58
llama_stack/providers/remote/inference/nvidia/NVIDIA.md
View file

@ -41,10 +41,10 @@ client.initialize()
### Create Completion
> Note on Completion API
>
> The hosted NVIDIA Llama NIMs (e.g., `meta-llama/Llama-3.1-8B-Instruct`) with ```NVIDIA_BASE_URL="https://integrate.api.nvidia.com"``` does not support the ```completion``` method, while the locally deployed NIM does.
The following example shows how to create a completion for an NVIDIA NIM.
> [!NOTE]
> The hosted NVIDIA Llama NIMs (for example ```meta-llama/Llama-3.1-8B-Instruct```) that have ```NVIDIA_BASE_URL="https://integrate.api.nvidia.com"``` do not support the ```completion``` method, while locally deployed NIMs do.
```python
response = client.inference.completion(
@ -60,6 +60,8 @@ print(f"Response: {response.content}")
### Create Chat Completion
The following example shows how to create a chat completion for an NVIDIA NIM.
```python
response = client.inference.chat_completion(
model_id="meta-llama/Llama-3.1-8B-Instruct",
@ -82,6 +84,9 @@ print(f"Response: {response.completion_message.content}")
```
### Tool Calling Example ###
The following example shows how to do tool calling for an NVIDIA NIM.
```python
from llama_stack.models.llama.datatypes import ToolDefinition, ToolParamDefinition
@ -117,6 +122,9 @@ if tool_response.completion_message.tool_calls:
```
### Structured Output Example
The following example shows how to do structured output for an NVIDIA NIM.
```python
from llama_stack.apis.inference import JsonSchemaResponseFormat, ResponseFormatType
@ -149,8 +157,10 @@ print(f"Structured Response: {structured_response.completion_message.content}")
```
### Create Embeddings
> Note on OpenAI embeddings compatibility
>
The following example shows how to create embeddings for an NVIDIA NIM.
> [!NOTE]
> NVIDIA asymmetric embedding models (e.g., `nvidia/llama-3.2-nv-embedqa-1b-v2`) require an `input_type` parameter not present in the standard OpenAI embeddings API. The NVIDIA Inference Adapter automatically sets `input_type="query"` when using the OpenAI-compatible embeddings endpoint for NVIDIA. For passage embeddings, use the `embeddings` API with `task_type="document"`.
```python
@ -161,3 +171,41 @@ response = client.inference.embeddings(
)
print(f"Embeddings: {response.embeddings}")
```
### Vision Language Models Example
The following example shows how to run vision inference by using an NVIDIA NIM.
```python
def load_image_as_base64(image_path):
with open(image_path, "rb") as image_file:
img_bytes = image_file.read()
return base64.b64encode(img_bytes).decode("utf-8")
image_path = {path_to_the_image}
demo_image_b64 = load_image_as_base64(image_path)
vlm_response = client.inference.chat_completion(
model_id="nvidia/vila",
messages=[
{
"role": "user",
"content": [
{
"type": "image",
"image": {
"data": demo_image_b64,
},
},
{
"type": "text",
"text": "Please describe what you see in this image in detail.",
},
],
}
],
)
print(f"VLM Response: {vlm_response.completion_message.content}")
```

4
llama_stack/providers/remote/inference/nvidia/models.py
View file

@ -55,6 +55,10 @@ MODEL_ENTRIES = [
"meta/llama-3.3-70b-instruct",
CoreModelId.llama3_3_70b_instruct.value,
),
ProviderModelEntry(
provider_model_id="nvidia/vila",
model_type=ModelType.llm,
),
# NeMo Retriever Text Embedding models -
#
# https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html

8
llama_stack/providers/remote/inference/ollama/ollama.py
View file

@ -118,10 +118,10 @@ class OllamaInferenceAdapter(
async def initialize(self) -> None:
logger.info(f"checking connectivity to Ollama at `{self.config.url}`...")
health_response = await self.health()
if health_response["status"] == HealthStatus.ERROR:
r = await self.health()
if r["status"] == HealthStatus.ERROR:
logger.warning(
"Ollama Server is not running, make sure to start it using `ollama serve` in a separate terminal"
f"Ollama Server is not running (message: {r['message']}). Make sure to start it using `ollama serve` in a separate terminal"
)
async def should_refresh_models(self) -> bool:
@ -156,7 +156,7 @@ class OllamaInferenceAdapter(
),
Model(
identifier="nomic-embed-text",
provider_resource_id="nomic-embed-text",
provider_resource_id="nomic-embed-text:latest",
provider_id=provider_id,
metadata={
"embedding_dimension": 768,

4
llama_stack/providers/remote/inference/watsonx/watsonx.py
View file

@ -7,8 +7,8 @@
from collections.abc import AsyncGenerator, AsyncIterator
from typing import Any
from ibm_watson_machine_learning.foundation_models import Model
from ibm_watson_machine_learning.metanames import GenTextParamsMetaNames as GenParams
from ibm_watsonx_ai.foundation_models import Model
from ibm_watsonx_ai.metanames import GenTextParamsMetaNames as GenParams
from openai import AsyncOpenAI
from llama_stack.apis.common.content_types import InterleavedContent, InterleavedContentItem

208
llama_stack/providers/remote/vector_io/pgvector/pgvector.py
View file

@ -4,6 +4,7 @@
# This source code is licensed under the terms described in the LICENSE file in
# the root directory of this source tree.
import heapq
from typing import Any
import psycopg2
@ -23,6 +24,9 @@ from llama_stack.apis.vector_io import (
)
from llama_stack.log import get_logger
from llama_stack.providers.datatypes import Api, VectorDBsProtocolPrivate
from llama_stack.providers.utils.inference.prompt_adapter import (
interleaved_content_as_str,
)
from llama_stack.providers.utils.kvstore import kvstore_impl
from llama_stack.providers.utils.kvstore.api import KVStore
from llama_stack.providers.utils.memory.openai_vector_store_mixin import OpenAIVectorStoreMixin
@ -31,6 +35,7 @@ from llama_stack.providers.utils.memory.vector_store import (
EmbeddingIndex,
VectorDBWithIndex,
)
from llama_stack.providers.utils.vector_io.vector_utils import WeightedInMemoryAggregator, sanitize_collection_name
from .config import PGVectorVectorIOConfig
@ -72,26 +77,64 @@ def load_models(cur, cls):
class PGVectorIndex(EmbeddingIndex):
def __init__(self, vector_db: VectorDB, dimension: int, conn, kvstore: KVStore | None = None):
# reference: https://github.com/pgvector/pgvector?tab=readme-ov-file#querying
PGVECTOR_DISTANCE_METRIC_TO_SEARCH_FUNCTION: dict[str, str] = {
"L2": "<->",
"L1": "<+>",
"COSINE": "<=>",
"INNER_PRODUCT": "<#>",
"HAMMING": "<~>",
"JACCARD": "<%>",
}
def __init__(
self,
vector_db: VectorDB,
dimension: int,
conn: psycopg2.extensions.connection,
kvstore: KVStore | None = None,
distance_metric: str = "COSINE",
):
self.vector_db = vector_db
self.dimension = dimension
self.conn = conn
with conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
self.kvstore = kvstore
self.check_distance_metric_availability(distance_metric)
self.distance_metric = distance_metric
self.table_name = None
async def initialize(self) -> None:
try:
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
# Sanitize the table name by replacing hyphens with underscores
# SQL doesn't allow hyphens in table names, and vector_db.identifier may contain hyphens
# when created with patterns like "test-vector-db-{uuid4()}"
sanitized_identifier = vector_db.identifier.replace("-", "_")
self.table_name = f"vector_store_{sanitized_identifier}"
self.kvstore = kvstore
sanitized_identifier = sanitize_collection_name(self.vector_db.identifier)
self.table_name = f"vs_{sanitized_identifier}"
cur.execute(
f"""
CREATE TABLE IF NOT EXISTS {self.table_name} (
id TEXT PRIMARY KEY,
document JSONB,
embedding vector({dimension})
embedding vector({self.dimension}),
content_text TEXT,
tokenized_content TSVECTOR
)
"""
)
# Create GIN index for full-text search performance
cur.execute(
f"""
CREATE INDEX IF NOT EXISTS {self.table_name}_content_gin_idx
ON {self.table_name} USING GIN(tokenized_content)
"""
)
except Exception as e:
log.exception(f"Error creating PGVectorIndex for vector_db: {self.vector_db.identifier}")
raise RuntimeError(f"Error creating PGVectorIndex for vector_db: {self.vector_db.identifier}") from e
async def add_chunks(self, chunks: list[Chunk], embeddings: NDArray):
assert len(chunks) == len(embeddings), (
f"Chunk length {len(chunks)} does not match embedding length {len(embeddings)}"
@ -99,29 +142,49 @@ class PGVectorIndex(EmbeddingIndex):
values = []
for i, chunk in enumerate(chunks):
content_text = interleaved_content_as_str(chunk.content)
values.append(
(
f"{chunk.chunk_id}",
Json(chunk.model_dump()),
embeddings[i].tolist(),
content_text,
content_text, # Pass content_text twice - once for content_text column, once for to_tsvector function. Eg. to_tsvector(content_text) = tokenized_content
)
)
query = sql.SQL(
f"""
INSERT INTO {self.table_name} (id, document, embedding)
INSERT INTO {self.table_name} (id, document, embedding, content_text, tokenized_content)
VALUES %s
ON CONFLICT (id) DO UPDATE SET embedding = EXCLUDED.embedding, document = EXCLUDED.document
ON CONFLICT (id) DO UPDATE SET
embedding = EXCLUDED.embedding,
document = EXCLUDED.document,
content_text = EXCLUDED.content_text,
tokenized_content = EXCLUDED.tokenized_content
"""
)
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
execute_values(cur, query, values, template="(%s, %s, %s::vector)")
execute_values(cur, query, values, template="(%s, %s, %s::vector, %s, to_tsvector('english', %s))")
async def query_vector(self, embedding: NDArray, k: int, score_threshold: float) -> QueryChunksResponse:
"""
Performs vector similarity search using PostgreSQL's search function. Default distance metric is COSINE.
Args:
embedding: The query embedding vector
k: Number of results to return
score_threshold: Minimum similarity score threshold
Returns:
QueryChunksResponse with combined results
"""
pgvector_search_function = self.get_pgvector_search_function()
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
cur.execute(
f"""
SELECT document, embedding <-> %s::vector AS distance
SELECT document, embedding {pgvector_search_function} %s::vector AS distance
FROM {self.table_name}
ORDER BY distance
LIMIT %s
@ -147,7 +210,40 @@ class PGVectorIndex(EmbeddingIndex):
k: int,
score_threshold: float,
) -> QueryChunksResponse:
raise NotImplementedError("Keyword search is not supported in PGVector")
"""
Performs keyword-based search using PostgreSQL's full-text search with ts_rank scoring.
Args:
query_string: The text query for keyword search
k: Number of results to return
score_threshold: Minimum similarity score threshold
Returns:
QueryChunksResponse with combined results
"""
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
# Use plainto_tsquery to handle user input safely and ts_rank for relevance scoring
cur.execute(
f"""
SELECT document, ts_rank(tokenized_content, plainto_tsquery('english', %s)) AS score
FROM {self.table_name}
WHERE tokenized_content @@ plainto_tsquery('english', %s)
ORDER BY score DESC
LIMIT %s
""",
(query_string, query_string, k),
)
results = cur.fetchall()
chunks = []
scores = []
for doc, score in results:
if score < score_threshold:
continue
chunks.append(Chunk(**doc))
scores.append(float(score))
return QueryChunksResponse(chunks=chunks, scores=scores)
async def query_hybrid(
self,
@ -158,7 +254,59 @@ class PGVectorIndex(EmbeddingIndex):
reranker_type: str,
reranker_params: dict[str, Any] | None = None,
) -> QueryChunksResponse:
raise NotImplementedError("Hybrid search is not supported in PGVector")
"""
Hybrid search combining vector similarity and keyword search using configurable reranking.
Args:
embedding: The query embedding vector
query_string: The text query for keyword search
k: Number of results to return
score_threshold: Minimum similarity score threshold
reranker_type: Type of reranker to use ("rrf" or "weighted")
reranker_params: Parameters for the reranker
Returns:
QueryChunksResponse with combined results
"""
if reranker_params is None:
reranker_params = {}
# Get results from both search methods
vector_response = await self.query_vector(embedding, k, score_threshold)
keyword_response = await self.query_keyword(query_string, k, score_threshold)
# Convert responses to score dictionaries using chunk_id
vector_scores = {
chunk.chunk_id: score for chunk, score in zip(vector_response.chunks, vector_response.scores, strict=False)
}
keyword_scores = {
chunk.chunk_id: score
for chunk, score in zip(keyword_response.chunks, keyword_response.scores, strict=False)
}
# Combine scores using the reranking utility
combined_scores = WeightedInMemoryAggregator.combine_search_results(
vector_scores, keyword_scores, reranker_type, reranker_params
)
# Efficient top-k selection because it only tracks the k best candidates it's seen so far
top_k_items = heapq.nlargest(k, combined_scores.items(), key=lambda x: x[1])
# Filter by score threshold
filtered_items = [(doc_id, score) for doc_id, score in top_k_items if score >= score_threshold]
# Create a map of chunk_id to chunk for both responses
chunk_map = {c.chunk_id: c for c in vector_response.chunks + keyword_response.chunks}
# Use the map to look up chunks by their IDs
chunks = []
scores = []
for doc_id, score in filtered_items:
if doc_id in chunk_map:
chunks.append(chunk_map[doc_id])
scores.append(score)
return QueryChunksResponse(chunks=chunks, scores=scores)
async def delete(self):
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
@ -170,6 +318,25 @@ class PGVectorIndex(EmbeddingIndex):
with self.conn.cursor(cursor_factory=psycopg2.extras.DictCursor) as cur:
cur.execute(f"DELETE FROM {self.table_name} WHERE id = ANY(%s)", (chunk_ids,))
def get_pgvector_search_function(self) -> str:
return self.PGVECTOR_DISTANCE_METRIC_TO_SEARCH_FUNCTION[self.distance_metric]
def check_distance_metric_availability(self, distance_metric: str) -> None:
"""Check if the distance metric is supported by PGVector.
Args:
distance_metric: The distance metric to check
Raises:
ValueError: If the distance metric is not supported
"""
if distance_metric not in self.PGVECTOR_DISTANCE_METRIC_TO_SEARCH_FUNCTION:
supported_metrics = list(self.PGVECTOR_DISTANCE_METRIC_TO_SEARCH_FUNCTION.keys())
raise ValueError(
f"Distance metric '{distance_metric}' is not supported by PGVector. "
f"Supported metrics are: {', '.join(supported_metrics)}"
)
class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtocolPrivate):
def __init__(
@ -185,8 +352,8 @@ class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoco
self.files_api = files_api
self.kvstore: KVStore | None = None
self.vector_db_store = None
self.openai_vector_store: dict[str, dict[str, Any]] = {}
self.metadatadata_collection_name = "openai_vector_stores_metadata"
self.openai_vector_stores: dict[str, dict[str, Any]] = {}
self.metadata_collection_name = "openai_vector_stores_metadata"
async def initialize(self) -> None:
log.info(f"Initializing PGVector memory adapter with config: {self.config}")
@ -233,9 +400,13 @@ class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoco
upsert_models(self.conn, [(vector_db.identifier, vector_db)])
# Create and cache the PGVector index table for the vector DB
pgvector_index = PGVectorIndex(
vector_db=vector_db, dimension=vector_db.embedding_dimension, conn=self.conn, kvstore=self.kvstore
)
await pgvector_index.initialize()
index = VectorDBWithIndex(
vector_db,
index=PGVectorIndex(vector_db, vector_db.embedding_dimension, self.conn, kvstore=self.kvstore),
index=pgvector_index,
inference_api=self.inference_api,
)
self.cache[vector_db.identifier] = index
@ -272,8 +443,15 @@ class PGVectorVectorIOAdapter(OpenAIVectorStoreMixin, VectorIO, VectorDBsProtoco
if vector_db_id in self.cache:
return self.cache[vector_db_id]
if self.vector_db_store is None:
raise VectorStoreNotFoundError(vector_db_id)
vector_db = await self.vector_db_store.get_vector_db(vector_db_id)
if not vector_db:
raise VectorStoreNotFoundError(vector_db_id)
index = PGVectorIndex(vector_db, vector_db.embedding_dimension, self.conn)
await index.initialize()
self.cache[vector_db_id] = VectorDBWithIndex(vector_db, index, self.inference_api)
return self.cache[vector_db_id]

10
llama_stack/providers/utils/memory/vector_store.py
View file

@ -294,12 +294,12 @@ class VectorDBWithIndex:
_validate_embedding(c.embedding, i, self.vector_db.embedding_dimension)
if chunks_to_embed:
resp = await self.inference_api.embeddings(
resp = await self.inference_api.openai_embeddings(
self.vector_db.embedding_model,
[c.content for c in chunks_to_embed],
)
for c, embedding in zip(chunks_to_embed, resp.embeddings, strict=False):
c.embedding = embedding
for c, data in zip(chunks_to_embed, resp.data, strict=False):
c.embedding = data.embedding
embeddings = np.array([c.embedding for c in chunks], dtype=np.float32)
await self.index.add_chunks(chunks, embeddings)
@ -334,8 +334,8 @@ class VectorDBWithIndex:
if mode == "keyword":
return await self.index.query_keyword(query_string, k, score_threshold)
embeddings_response = await self.inference_api.embeddings(self.vector_db.embedding_model, [query_string])
query_vector = np.array(embeddings_response.embeddings[0], dtype=np.float32)
embeddings_response = await self.inference_api.openai_embeddings(self.vector_db.embedding_model, [query_string])
query_vector = np.array(embeddings_response.data[0].embedding, dtype=np.float32)
if mode == "hybrid":
return await self.index.query_hybrid(
query_vector, query_string, k, score_threshold, reranker_type, reranker_params

31
llama_stack/providers/utils/sqlstore/sqlalchemy_sqlstore.py
View file

@ -23,6 +23,7 @@ from sqlalchemy import (
)
from sqlalchemy.ext.asyncio import async_sessionmaker, create_async_engine
from sqlalchemy.ext.asyncio.engine import AsyncEngine
from sqlalchemy.sql.elements import ColumnElement
from llama_stack.apis.common.responses import PaginatedResponse
from llama_stack.log import get_logger
@ -43,6 +44,30 @@ TYPE_MAPPING: dict[ColumnType, Any] = {
}
def _build_where_expr(column: ColumnElement, value: Any) -> ColumnElement:
"""Return a SQLAlchemy expression for a where condition.
`value` may be a simple scalar (equality) or a mapping like {">": 123}.
The returned expression is a SQLAlchemy ColumnElement usable in query.where(...).
"""
if isinstance(value, Mapping):
if len(value) != 1:
raise ValueError(f"Operator mapping must have a single operator, got: {value}")
op, operand = next(iter(value.items()))
if op == "==" or op == "=":
return column == operand
if op == ">":
return column > operand
if op == "<":
return column < operand
if op == ">=":
return column >= operand
if op == "<=":
return column <= operand
raise ValueError(f"Unsupported operator '{op}' in where mapping")
return column == value
class SqlAlchemySqlStoreImpl(SqlStore):
def __init__(self, config: SqlAlchemySqlStoreConfig):
self.config = config
@ -111,7 +136,7 @@ class SqlAlchemySqlStoreImpl(SqlStore):
if where:
for key, value in where.items():
query = query.where(table_obj.c[key] == value)
query = query.where(_build_where_expr(table_obj.c[key], value))
if where_sql:
query = query.where(text(where_sql))
@ -222,7 +247,7 @@ class SqlAlchemySqlStoreImpl(SqlStore):
async with self.async_session() as session:
stmt = self.metadata.tables[table].update()
for key, value in where.items():
stmt = stmt.where(self.metadata.tables[table].c[key] == value)
stmt = stmt.where(_build_where_expr(self.metadata.tables[table].c[key], value))
await session.execute(stmt, data)
await session.commit()
@ -233,7 +258,7 @@ class SqlAlchemySqlStoreImpl(SqlStore):
async with self.async_session() as session:
stmt = self.metadata.tables[table].delete()
for key, value in where.items():
stmt = stmt.where(self.metadata.tables[table].c[key] == value)
stmt = stmt.where(_build_where_expr(self.metadata.tables[table].c[key], value))
await session.execute(stmt)
await session.commit()

119
llama_stack/providers/utils/vector_io/vector_utils.py
View file

@ -37,3 +37,122 @@ def sanitize_collection_name(name: str, weaviate_format=False) -> str:
else:
s = proper_case(re.sub(r"[^a-zA-Z0-9]", "", name))
return s
class WeightedInMemoryAggregator:
@staticmethod
def _normalize_scores(scores: dict[str, float]) -> dict[str, float]:
"""
Normalize scores to 0-1 range using min-max normalization.
Args:
scores: dictionary of scores with document IDs as keys and scores as values
Returns:
Normalized scores with document IDs as keys and normalized scores as values
"""
if not scores:
return {}
min_score, max_score = min(scores.values()), max(scores.values())
score_range = max_score - min_score
if score_range > 0:
return {doc_id: (score - min_score) / score_range for doc_id, score in scores.items()}
return dict.fromkeys(scores, 1.0)
@staticmethod
def weighted_rerank(
vector_scores: dict[str, float],
keyword_scores: dict[str, float],
alpha: float = 0.5,
) -> dict[str, float]:
"""
Rerank via weighted average of scores.
Args:
vector_scores: scores from vector search
keyword_scores: scores from keyword search
alpha: weight factor between 0 and 1 (default: 0.5)
0 = keyword only, 1 = vector only, 0.5 = equal weight
Returns:
All unique document IDs with weighted combined scores
"""
all_ids = set(vector_scores.keys()) | set(keyword_scores.keys())
normalized_vector_scores = WeightedInMemoryAggregator._normalize_scores(vector_scores)
normalized_keyword_scores = WeightedInMemoryAggregator._normalize_scores(keyword_scores)
# Weighted formula: score = (1-alpha) * keyword_score + alpha * vector_score
# alpha=0 means keyword only, alpha=1 means vector only
return {
doc_id: ((1 - alpha) * normalized_keyword_scores.get(doc_id, 0.0))
+ (alpha * normalized_vector_scores.get(doc_id, 0.0))
for doc_id in all_ids
}
@staticmethod
def rrf_rerank(
vector_scores: dict[str, float],
keyword_scores: dict[str, float],
impact_factor: float = 60.0,
) -> dict[str, float]:
"""
Rerank via Reciprocal Rank Fusion.
Args:
vector_scores: scores from vector search
keyword_scores: scores from keyword search
impact_factor: impact factor for RRF (default: 60.0)
Returns:
All unique document IDs with RRF combined scores
"""
# Convert scores to ranks
vector_ranks = {
doc_id: i + 1
for i, (doc_id, _) in enumerate(sorted(vector_scores.items(), key=lambda x: x[1], reverse=True))
}
keyword_ranks = {
doc_id: i + 1
for i, (doc_id, _) in enumerate(sorted(keyword_scores.items(), key=lambda x: x[1], reverse=True))
}
all_ids = set(vector_scores.keys()) | set(keyword_scores.keys())
rrf_scores = {}
for doc_id in all_ids:
vector_rank = vector_ranks.get(doc_id, float("inf"))
keyword_rank = keyword_ranks.get(doc_id, float("inf"))
# RRF formula: score = 1/(k + r) where k is impact_factor (default: 60.0) and r is the rank
rrf_scores[doc_id] = (1.0 / (impact_factor + vector_rank)) + (1.0 / (impact_factor + keyword_rank))
return rrf_scores
@staticmethod
def combine_search_results(
vector_scores: dict[str, float],
keyword_scores: dict[str, float],
reranker_type: str = "rrf",
reranker_params: dict[str, float] | None = None,
) -> dict[str, float]:
"""
Combine vector and keyword search results using specified reranking strategy.
Args:
vector_scores: scores from vector search
keyword_scores: scores from keyword search
reranker_type: type of reranker to use (default: RERANKER_TYPE_RRF)
reranker_params: parameters for the reranker
Returns:
All unique document IDs with combined scores
"""
if reranker_params is None:
reranker_params = {}
if reranker_type == "weighted":
alpha = reranker_params.get("alpha", 0.5)
return WeightedInMemoryAggregator.weighted_rerank(vector_scores, keyword_scores, alpha)
else:
# Default to RRF for None, RRF, or any unknown types
impact_factor = reranker_params.get("impact_factor", 60.0)
return WeightedInMemoryAggregator.rrf_rerank(vector_scores, keyword_scores, impact_factor)

132
llama_stack/testing/inference_recorder.py
View file

@ -30,6 +30,9 @@ from openai.types.completion_choice import CompletionChoice
CompletionChoice.model_fields["finish_reason"].annotation = Literal["stop", "length", "content_filter"] | None
CompletionChoice.model_rebuild()
REPO_ROOT = Path(__file__).parent.parent.parent
DEFAULT_STORAGE_DIR = REPO_ROOT / "tests/integration/recordings"
class InferenceMode(StrEnum):
LIVE = "live"
@ -51,7 +54,7 @@ def normalize_request(method: str, url: str, headers: dict[str, Any], body: dict
def get_inference_mode() -> InferenceMode:
return InferenceMode(os.environ.get("LLAMA_STACK_TEST_INFERENCE_MODE", "live").lower())
return InferenceMode(os.environ.get("LLAMA_STACK_TEST_INFERENCE_MODE", "replay").lower())
def setup_inference_recording():
@ -60,28 +63,18 @@ def setup_inference_recording():
to increase their reliability and reduce reliance on expensive, external services.
Currently, this is only supported for OpenAI and Ollama clients. These should cover the vast majority of use cases.
Calls to the /models endpoint are not currently trapped. We probably need to add support for this.
Two environment variables are required:
- LLAMA_STACK_TEST_INFERENCE_MODE: The mode to run in. Must be 'live', 'record', or 'replay'.
- LLAMA_STACK_TEST_RECORDING_DIR: The directory to store the recordings in.
Two environment variables are supported:
- LLAMA_STACK_TEST_INFERENCE_MODE: The mode to run in. Must be 'live', 'record', or 'replay'. Default is 'replay'.
- LLAMA_STACK_TEST_RECORDING_DIR: The directory to store the recordings in. Default is 'tests/integration/recordings'.
The recordings are stored in a SQLite database and a JSON file for each request. The SQLite database is used to
quickly find the correct recording for a given request. The JSON files are used to store the request and response
bodies.
The recordings are stored as JSON files.
"""
mode = get_inference_mode()
if mode not in InferenceMode:
raise ValueError(f"Invalid LLAMA_STACK_TEST_INFERENCE_MODE: {mode}. Must be 'live', 'record', or 'replay'")
if mode == InferenceMode.LIVE:
return None
if "LLAMA_STACK_TEST_RECORDING_DIR" not in os.environ:
raise ValueError("LLAMA_STACK_TEST_RECORDING_DIR must be set for recording or replaying")
storage_dir = os.environ["LLAMA_STACK_TEST_RECORDING_DIR"]
storage_dir = os.environ.get("LLAMA_STACK_TEST_RECORDING_DIR", DEFAULT_STORAGE_DIR)
return inference_recording(mode=mode, storage_dir=storage_dir)
@ -134,8 +127,8 @@ class ResponseStorage:
def store_recording(self, request_hash: str, request: dict[str, Any], response: dict[str, Any]):
"""Store a request/response pair."""
# Generate unique response filename
response_file = f"{request_hash[:12]}.json"
response_path = self.responses_dir / response_file
short_hash = request_hash[:12]
response_file = f"{short_hash}.json"
# Serialize response body if needed
serialized_response = dict(response)
@ -147,6 +140,14 @@ class ResponseStorage:
# Handle single response
serialized_response["body"] = _serialize_response(serialized_response["body"])
# If this is an Ollama /api/tags recording, include models digest in filename to distinguish variants
endpoint = request.get("endpoint")
if endpoint in ("/api/tags", "/v1/models"):
digest = _model_identifiers_digest(endpoint, response)
response_file = f"models-{short_hash}-{digest}.json"
response_path = self.responses_dir / response_file
# Save response to JSON file
with open(response_path, "w") as f:
json.dump({"request": request, "response": serialized_response}, f, indent=2)
@ -161,6 +162,17 @@ class ResponseStorage:
if not response_path.exists():
return None
return _recording_from_file(response_path)
def _model_list_responses(self, short_hash: str) -> list[dict[str, Any]]:
results: list[dict[str, Any]] = []
for path in self.responses_dir.glob(f"models-{short_hash}-*.json"):
data = _recording_from_file(path)
results.append(data)
return results
def _recording_from_file(response_path) -> dict[str, Any]:
with open(response_path) as f:
data = json.load(f)
@ -176,6 +188,61 @@ class ResponseStorage:
return cast(dict[str, Any], data)
def _model_identifiers_digest(endpoint: str, response: dict[str, Any]) -> str:
def _extract_model_identifiers():
"""Extract a stable set of identifiers for model-list endpoints.
Supported endpoints:
- '/api/tags' (Ollama): response body has 'models': [ { name/model/digest/id/... }, ... ]
- '/v1/models' (OpenAI): response body has 'data': [ { id: ... }, ... ]
Returns a list of unique identifiers or None if structure doesn't match.
"""
body = response["body"]
if endpoint == "/api/tags":
items = body.get("models")
idents = [m.model for m in items]
else:
items = body.get("data")
idents = [m.id for m in items]
return sorted(set(idents))
identifiers = _extract_model_identifiers()
return hashlib.sha1(("|".join(identifiers)).encode("utf-8")).hexdigest()[:8]
def _combine_model_list_responses(endpoint: str, records: list[dict[str, Any]]) -> dict[str, Any] | None:
"""Return a single, unioned recording for supported model-list endpoints."""
seen: dict[str, dict[str, Any]] = {}
for rec in records:
body = rec["response"]["body"]
if endpoint == "/api/tags":
items = body.models
elif endpoint == "/v1/models":
items = body.data
else:
items = []
for m in items:
if endpoint == "/v1/models":
key = m.id
else:
key = m.model
seen[key] = m
ordered = [seen[k] for k in sorted(seen.keys())]
canonical = records[0]
canonical_req = canonical.get("request", {})
if isinstance(canonical_req, dict):
canonical_req["endpoint"] = endpoint
if endpoint == "/v1/models":
body = {"data": ordered, "object": "list"}
else:
from ollama import ListResponse
body = ListResponse(models=ordered)
return {"request": canonical_req, "response": {"body": body, "is_streaming": False}}
async def _patched_inference_method(original_method, self, client_type, endpoint, *args, **kwargs):
global _current_mode, _current_storage
@ -195,8 +262,6 @@ async def _patched_inference_method(original_method, self, client_type, endpoint
raise ValueError(f"Unknown client type: {client_type}")
url = base_url.rstrip("/") + endpoint
# Normalize request for matching
method = "POST"
headers = {}
body = kwargs
@ -204,6 +269,11 @@ async def _patched_inference_method(original_method, self, client_type, endpoint
request_hash = normalize_request(method, url, headers, body)
if _current_mode == InferenceMode.REPLAY:
# Special handling for model-list endpoints: return union of all responses
if endpoint in ("/api/tags", "/v1/models"):
records = _current_storage._model_list_responses(request_hash[:12])
recording = _combine_model_list_responses(endpoint, records)
else:
recording = _current_storage.find_recording(request_hash)
if recording:
response_body = recording["response"]["body"]
@ -274,12 +344,14 @@ def patch_inference_clients():
from openai.resources.chat.completions import AsyncCompletions as AsyncChatCompletions
from openai.resources.completions import AsyncCompletions
from openai.resources.embeddings import AsyncEmbeddings
from openai.resources.models import AsyncModels
# Store original methods for both OpenAI and Ollama clients
_original_methods = {
"chat_completions_create": AsyncChatCompletions.create,
"completions_create": AsyncCompletions.create,
"embeddings_create": AsyncEmbeddings.create,
"models_list": AsyncModels.list,
"ollama_generate": OllamaAsyncClient.generate,
"ollama_chat": OllamaAsyncClient.chat,
"ollama_embed": OllamaAsyncClient.embed,
@ -304,10 +376,16 @@ def patch_inference_clients():
_original_methods["embeddings_create"], self, "openai", "/v1/embeddings", *args, **kwargs
)
async def patched_models_list(self, *args, **kwargs):
return await _patched_inference_method(
_original_methods["models_list"], self, "openai", "/v1/models", *args, **kwargs
)
# Apply OpenAI patches
AsyncChatCompletions.create = patched_chat_completions_create
AsyncCompletions.create = patched_completions_create
AsyncEmbeddings.create = patched_embeddings_create
AsyncModels.list = patched_models_list
# Create patched methods for Ollama client
async def patched_ollama_generate(self, *args, **kwargs):
@ -361,11 +439,13 @@ def unpatch_inference_clients():
from openai.resources.chat.completions import AsyncCompletions as AsyncChatCompletions
from openai.resources.completions import AsyncCompletions
from openai.resources.embeddings import AsyncEmbeddings
from openai.resources.models import AsyncModels
# Restore OpenAI client methods
AsyncChatCompletions.create = _original_methods["chat_completions_create"]
AsyncCompletions.create = _original_methods["completions_create"]
AsyncEmbeddings.create = _original_methods["embeddings_create"]
AsyncModels.list = _original_methods["models_list"]
# Restore Ollama client methods if they were patched
OllamaAsyncClient.generate = _original_methods["ollama_generate"]
@ -379,16 +459,10 @@ def unpatch_inference_clients():
@contextmanager
def inference_recording(mode: str = "live", storage_dir: str | Path | None = None) -> Generator[None, None, None]:
def inference_recording(mode: str, storage_dir: str | Path | None = None) -> Generator[None, None, None]:
"""Context manager for inference recording/replaying."""
global _current_mode, _current_storage
# Set defaults
if storage_dir is None:
storage_dir_path = Path.home() / ".llama" / "recordings"
else:
storage_dir_path = Path(storage_dir)
# Store previous state
prev_mode = _current_mode
prev_storage = _current_storage
@ -397,7 +471,9 @@ def inference_recording(mode: str = "live", storage_dir: str | Path | None = Non
_current_mode = mode
if mode in ["record", "replay"]:
_current_storage = ResponseStorage(storage_dir_path)
if storage_dir is None:
raise ValueError("storage_dir is required for record and replay modes")
_current_storage = ResponseStorage(Path(storage_dir))
patch_inference_clients()
yield

305
llama_stack/ui/package-lock.json generated
View file

@ -14,11 +14,11 @@
"@radix-ui/react-select": "^2.2.5",
"@radix-ui/react-separator": "^1.1.7",
"@radix-ui/react-slot": "^1.2.3",
"@radix-ui/react-tooltip": "^1.2.6",
"@radix-ui/react-tooltip": "^1.2.8",
"class-variance-authority": "^0.7.1",
"clsx": "^2.1.1",
"framer-motion": "^11.18.2",
"llama-stack-client": "^0.2.19",
"framer-motion": "^12.23.12",
"llama-stack-client": "^0.2.20",
"lucide-react": "^0.510.0",
"next": "15.3.3",
"next-auth": "^4.24.11",
@ -39,16 +39,16 @@
"@testing-library/jest-dom": "^6.8.0",
"@testing-library/react": "^16.3.0",
"@types/jest": "^29.5.14",
"@types/node": "^20",
"@types/node": "^24",
"@types/react": "^19",
"@types/react-dom": "^19",
"eslint": "^9",
"eslint-config-next": "15.3.2",
"eslint-config-next": "15.5.2",
"eslint-config-prettier": "^10.1.8",
"eslint-plugin-prettier": "^5.5.4",
"jest": "^29.7.0",
"jest-environment-jsdom": "^29.7.0",
"prettier": "3.5.3",
"prettier": "3.6.2",
"tailwindcss": "^4",
"ts-node": "^10.9.2",
"tw-animate-css": "^1.2.9",
@ -1854,9 +1854,9 @@
"integrity": "sha512-OdiMrzCl2Xi0VTjiQQUK0Xh7bJHnOuET2s+3V+Y40WJBAXrJeGA3f+I8MZJ/YQ3mVGi5XGR1L66oFlgqXhQ4Vw=="
},
"node_modules/@next/eslint-plugin-next": {
"version": "15.3.2",
"resolved": "https://registry.npmjs.org/@next/eslint-plugin-next/-/eslint-plugin-next-15.3.2.tgz",
"integrity": "sha512-ijVRTXBgnHT33aWnDtmlG+LJD+5vhc9AKTJPquGG5NKXjpKNjc62woIhFtrAcWdBobt8kqjCoaJ0q6sDQoX7aQ==",
"version": "15.5.2",
"resolved": "https://registry.npmjs.org/@next/eslint-plugin-next/-/eslint-plugin-next-15.5.2.tgz",
"integrity": "sha512-lkLrRVxcftuOsJNhWatf1P2hNVfh98k/omQHrCEPPriUypR6RcS13IvLdIrEvkm9AH2Nu2YpR5vLqBuy6twH3Q==",
"dev": true,
"license": "MIT",
"dependencies": {
@ -2861,29 +2861,6 @@
}
}
},
"node_modules/@radix-ui/react-select/node_modules/@radix-ui/react-visually-hidden": {
"version": "1.2.3",
"resolved": "https://registry.npmjs.org/@radix-ui/react-visually-hidden/-/react-visually-hidden-1.2.3.tgz",
"integrity": "sha512-pzJq12tEaaIhqjbzpCuv/OypJY/BPavOofm+dbab+MHLajy277+1lLm6JFcGgF5eskJ6mquGirhXY2GD/8u8Ug==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-primitive": "2.1.3"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-separator": {
"version": "1.1.7",
"resolved": "https://registry.npmjs.org/@radix-ui/react-separator/-/react-separator-1.1.7.tgz",
@ -2949,23 +2926,23 @@
}
},
"node_modules/@radix-ui/react-tooltip": {
"version": "1.2.6",
"resolved": "https://registry.npmjs.org/@radix-ui/react-tooltip/-/react-tooltip-1.2.6.tgz",
"integrity": "sha512-zYb+9dc9tkoN2JjBDIIPLQtk3gGyz8FMKoqYTb8EMVQ5a5hBcdHPECrsZVI4NpPAUOixhkoqg7Hj5ry5USowfA==",
"version": "1.2.8",
"resolved": "https://registry.npmjs.org/@radix-ui/react-tooltip/-/react-tooltip-1.2.8.tgz",
"integrity": "sha512-tY7sVt1yL9ozIxvmbtN5qtmH2krXcBCfjEiCgKGLqunJHvgvZG2Pcl2oQ3kbcZARb1BGEHdkLzcYGO8ynVlieg==",
"license": "MIT",
"dependencies": {
"@radix-ui/primitive": "1.1.2",
"@radix-ui/primitive": "1.1.3",
"@radix-ui/react-compose-refs": "1.1.2",
"@radix-ui/react-context": "1.1.2",
"@radix-ui/react-dismissable-layer": "1.1.9",
"@radix-ui/react-dismissable-layer": "1.1.11",
"@radix-ui/react-id": "1.1.1",
"@radix-ui/react-popper": "1.2.6",
"@radix-ui/react-portal": "1.1.8",
"@radix-ui/react-presence": "1.1.4",
"@radix-ui/react-primitive": "2.1.2",
"@radix-ui/react-slot": "1.2.2",
"@radix-ui/react-popper": "1.2.8",
"@radix-ui/react-portal": "1.1.9",
"@radix-ui/react-presence": "1.1.5",
"@radix-ui/react-primitive": "2.1.3",
"@radix-ui/react-slot": "1.2.3",
"@radix-ui/react-use-controllable-state": "1.2.2",
"@radix-ui/react-visually-hidden": "1.2.2"
"@radix-ui/react-visually-hidden": "1.2.3"
},
"peerDependencies": {
"@types/react": "*",
@ -2982,21 +2959,162 @@
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-slot": {
"version": "1.2.2",
"resolved": "https://registry.npmjs.org/@radix-ui/react-slot/-/react-slot-1.2.2.tgz",
"integrity": "sha512-y7TBO4xN4Y94FvcWIOIh18fM4R1A8S4q1jhoz4PNzOoHsFcN8pogcFmZrTYAm4F9VRUrWP/Mw7xSKybIeRI+CQ==",
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/primitive": {
"version": "1.1.3",
"resolved": "https://registry.npmjs.org/@radix-ui/primitive/-/primitive-1.1.3.tgz",
"integrity": "sha512-JTF99U/6XIjCBo0wqkU5sK10glYe27MRRsfwoiq5zzOEZLHU3A3KCMa5X/azekYRCJ0HlwI0crAXS/5dEHTzDg==",
"license": "MIT"
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-arrow": {
"version": "1.1.7",
"resolved": "https://registry.npmjs.org/@radix-ui/react-arrow/-/react-arrow-1.1.7.tgz",
"integrity": "sha512-F+M1tLhO+mlQaOWspE8Wstg+z6PwxwRd8oQ8IXceWz92kfAmalTRf0EjrouQeo7QssEPfCn05B4Ihs1K9WQ/7w==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-compose-refs": "1.1.2"
"@radix-ui/react-primitive": "2.1.3"
},
"peerDependencies": {
"@types/react": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-dismissable-layer": {
"version": "1.1.11",
"resolved": "https://registry.npmjs.org/@radix-ui/react-dismissable-layer/-/react-dismissable-layer-1.1.11.tgz",
"integrity": "sha512-Nqcp+t5cTB8BinFkZgXiMJniQH0PsUt2k51FUhbdfeKvc4ACcG2uQniY/8+h1Yv6Kza4Q7lD7PQV0z0oicE0Mg==",
"license": "MIT",
"dependencies": {
"@radix-ui/primitive": "1.1.3",
"@radix-ui/react-compose-refs": "1.1.2",
"@radix-ui/react-primitive": "2.1.3",
"@radix-ui/react-use-callback-ref": "1.1.1",
"@radix-ui/react-use-escape-keydown": "1.1.1"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-popper": {
"version": "1.2.8",
"resolved": "https://registry.npmjs.org/@radix-ui/react-popper/-/react-popper-1.2.8.tgz",
"integrity": "sha512-0NJQ4LFFUuWkE7Oxf0htBKS6zLkkjBH+hM1uk7Ng705ReR8m/uelduy1DBo0PyBXPKVnBA6YBlU94MBGXrSBCw==",
"license": "MIT",
"dependencies": {
"@floating-ui/react-dom": "^2.0.0",
"@radix-ui/react-arrow": "1.1.7",
"@radix-ui/react-compose-refs": "1.1.2",
"@radix-ui/react-context": "1.1.2",
"@radix-ui/react-primitive": "2.1.3",
"@radix-ui/react-use-callback-ref": "1.1.1",
"@radix-ui/react-use-layout-effect": "1.1.1",
"@radix-ui/react-use-rect": "1.1.1",
"@radix-ui/react-use-size": "1.1.1",
"@radix-ui/rect": "1.1.1"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-portal": {
"version": "1.1.9",
"resolved": "https://registry.npmjs.org/@radix-ui/react-portal/-/react-portal-1.1.9.tgz",
"integrity": "sha512-bpIxvq03if6UNwXZ+HTK71JLh4APvnXntDc6XOX8UVq4XQOVl7lwok0AvIl+b8zgCw3fSaVTZMpAPPagXbKmHQ==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-primitive": "2.1.3",
"@radix-ui/react-use-layout-effect": "1.1.1"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-presence": {
"version": "1.1.5",
"resolved": "https://registry.npmjs.org/@radix-ui/react-presence/-/react-presence-1.1.5.tgz",
"integrity": "sha512-/jfEwNDdQVBCNvjkGit4h6pMOzq8bHkopq458dPt2lMjx+eBQUohZNG9A7DtO/O5ukSbxuaNGXMjHicgwy6rQQ==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-compose-refs": "1.1.2",
"@radix-ui/react-use-layout-effect": "1.1.1"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-tooltip/node_modules/@radix-ui/react-primitive": {
"version": "2.1.3",
"resolved": "https://registry.npmjs.org/@radix-ui/react-primitive/-/react-primitive-2.1.3.tgz",
"integrity": "sha512-m9gTwRkhy2lvCPe6QJp4d3G1TYEUHn/FzJUtq9MjH46an1wJU+GdoGC5VLof8RX8Ft/DlpshApkhswDLZzHIcQ==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-slot": "1.2.3"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
@ -3137,12 +3255,35 @@
}
},
"node_modules/@radix-ui/react-visually-hidden": {
"version": "1.2.2",
"resolved": "https://registry.npmjs.org/@radix-ui/react-visually-hidden/-/react-visually-hidden-1.2.2.tgz",
"integrity": "sha512-ORCmRUbNiZIv6uV5mhFrhsIKw4UX/N3syZtyqvry61tbGm4JlgQuSn0hk5TwCARsCjkcnuRkSdCE3xfb+ADHew==",
"version": "1.2.3",
"resolved": "https://registry.npmjs.org/@radix-ui/react-visually-hidden/-/react-visually-hidden-1.2.3.tgz",
"integrity": "sha512-pzJq12tEaaIhqjbzpCuv/OypJY/BPavOofm+dbab+MHLajy277+1lLm6JFcGgF5eskJ6mquGirhXY2GD/8u8Ug==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-primitive": "2.1.2"
"@radix-ui/react-primitive": "2.1.3"
},
"peerDependencies": {
"@types/react": "*",
"@types/react-dom": "*",
"react": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc",
"react-dom": "^16.8 || ^17.0 || ^18.0 || ^19.0 || ^19.0.0-rc"
},
"peerDependenciesMeta": {
"@types/react": {
"optional": true
},
"@types/react-dom": {
"optional": true
}
}
},
"node_modules/@radix-ui/react-visually-hidden/node_modules/@radix-ui/react-primitive": {
"version": "2.1.3",
"resolved": "https://registry.npmjs.org/@radix-ui/react-primitive/-/react-primitive-2.1.3.tgz",
"integrity": "sha512-m9gTwRkhy2lvCPe6QJp4d3G1TYEUHn/FzJUtq9MjH46an1wJU+GdoGC5VLof8RX8Ft/DlpshApkhswDLZzHIcQ==",
"license": "MIT",
"dependencies": {
"@radix-ui/react-slot": "1.2.3"
},
"peerDependencies": {
"@types/react": "*",
@ -3910,12 +4051,12 @@
"license": "MIT"
},
"node_modules/@types/node": {
"version": "20.17.47",
"resolved": "https://registry.npmjs.org/@types/node/-/node-20.17.47.tgz",
"integrity": "sha512-3dLX0Upo1v7RvUimvxLeXqwrfyKxUINk0EAM83swP2mlSUcwV73sZy8XhNz8bcZ3VbsfQyC/y6jRdL5tgCNpDQ==",
"version": "24.3.0",
"resolved": "https://registry.npmjs.org/@types/node/-/node-24.3.0.tgz",
"integrity": "sha512-aPTXCrfwnDLj4VvXrm+UUCQjNEvJgNA8s5F1cvwQU+3KNltTOkBm1j30uNLyqqPNe7gE3KFzImYoZEfLhp4Yow==",
"license": "MIT",
"dependencies": {
"undici-types": "~6.19.2"
"undici-types": "~7.10.0"
}
},
"node_modules/@types/node-fetch": {
@ -6433,13 +6574,13 @@
}
},
"node_modules/eslint-config-next": {
"version": "15.3.2",
"resolved": "https://registry.npmjs.org/eslint-config-next/-/eslint-config-next-15.3.2.tgz",
"integrity": "sha512-FerU4DYccO4FgeYFFglz0SnaKRe1ejXQrDb8kWUkTAg036YWi+jUsgg4sIGNCDhAsDITsZaL4MzBWKB6f4G1Dg==",
"version": "15.5.2",
"resolved": "https://registry.npmjs.org/eslint-config-next/-/eslint-config-next-15.5.2.tgz",
"integrity": "sha512-3hPZghsLupMxxZ2ggjIIrat/bPniM2yRpsVPVM40rp8ZMzKWOJp2CGWn7+EzoV2ddkUr5fxNfHpF+wU1hGt/3g==",
"dev": true,
"license": "MIT",
"dependencies": {
"@next/eslint-plugin-next": "15.3.2",
"@next/eslint-plugin-next": "15.5.2",
"@rushstack/eslint-patch": "^1.10.3",
"@typescript-eslint/eslint-plugin": "^5.4.2 || ^6.0.0 || ^7.0.0 || ^8.0.0",
"@typescript-eslint/parser": "^5.4.2 || ^6.0.0 || ^7.0.0 || ^8.0.0",
@ -7268,13 +7409,13 @@
}
},
"node_modules/framer-motion": {
"version": "11.18.2",
"resolved": "https://registry.npmjs.org/framer-motion/-/framer-motion-11.18.2.tgz",
"integrity": "sha512-5F5Och7wrvtLVElIpclDT0CBzMVg3dL22B64aZwHtsIY8RB4mXICLrkajK4G9R+ieSAGcgrLeae2SeUTg2pr6w==",
"version": "12.23.12",
"resolved": "https://registry.npmjs.org/framer-motion/-/framer-motion-12.23.12.tgz",
"integrity": "sha512-6e78rdVtnBvlEVgu6eFEAgG9v3wLnYEboM8I5O5EXvfKC8gxGQB8wXJdhkMy10iVcn05jl6CNw7/HTsTCfwcWg==",
"license": "MIT",
"dependencies": {
"motion-dom": "^11.18.1",
"motion-utils": "^11.18.1",
"motion-dom": "^12.23.12",
"motion-utils": "^12.23.6",
"tslib": "^2.4.0"
},
"peerDependencies": {
@ -10006,9 +10147,9 @@
"license": "MIT"
},
"node_modules/llama-stack-client": {
"version": "0.2.19",
"resolved": "https://registry.npmjs.org/llama-stack-client/-/llama-stack-client-0.2.19.tgz",
"integrity": "sha512-sDuAhUdEGlERZ3jlMUzPXcQTgMv/pGbDrPX0ifbE5S+gr7Q+7ohuQYrIXe+hXgIipFjq+y4b2c5laZ76tmAyEA==",
"version": "0.2.20",
"resolved": "https://registry.npmjs.org/llama-stack-client/-/llama-stack-client-0.2.20.tgz",
"integrity": "sha512-1vD5nizTX5JEW8TADxKgy/P1W8YZoPSpdnmfxbdYbWgpQ3BWtbvLS6jmDk7VwVA5fRC4895VfHsRDfS1liHarw==",
"license": "MIT",
"dependencies": {
"@types/node": "^18.11.18",
@ -11184,18 +11325,18 @@
}
},
"node_modules/motion-dom": {
"version": "11.18.1",
"resolved": "https://registry.npmjs.org/motion-dom/-/motion-dom-11.18.1.tgz",
"integrity": "sha512-g76KvA001z+atjfxczdRtw/RXOM3OMSdd1f4DL77qCTF/+avrRJiawSG4yDibEQ215sr9kpinSlX2pCTJ9zbhw==",
"version": "12.23.12",
"resolved": "https://registry.npmjs.org/motion-dom/-/motion-dom-12.23.12.tgz",
"integrity": "sha512-RcR4fvMCTESQBD/uKQe49D5RUeDOokkGRmz4ceaJKDBgHYtZtntC/s2vLvY38gqGaytinij/yi3hMcWVcEF5Kw==",
"license": "MIT",
"dependencies": {
"motion-utils": "^11.18.1"
"motion-utils": "^12.23.6"
}
},
"node_modules/motion-utils": {
"version": "11.18.1",
"resolved": "https://registry.npmjs.org/motion-utils/-/motion-utils-11.18.1.tgz",
"integrity": "sha512-49Kt+HKjtbJKLtgO/LKj9Ld+6vw9BjH5d9sc40R/kVyH8GLAXgT42M2NnuPcJNuA3s9ZfZBUcwIgpmZWGEE+hA==",
"version": "12.23.6",
"resolved": "https://registry.npmjs.org/motion-utils/-/motion-utils-12.23.6.tgz",
"integrity": "sha512-eAWoPgr4eFEOFfg2WjIsMoqJTW6Z8MTUCgn/GZ3VRpClWBdnbjryiA3ZSNLyxCTmCQx4RmYX6jX1iWHbenUPNQ==",
"license": "MIT"
},
"node_modules/ms": {
@ -12083,9 +12224,9 @@
}
},
"node_modules/prettier": {
"version": "3.5.3",
"resolved": "https://registry.npmjs.org/prettier/-/prettier-3.5.3.tgz",
"integrity": "sha512-QQtaxnoDJeAkDvDKWCLiwIXkTgRhwYDEQCghU9Z6q03iyek/rxRh/2lC3HB7P8sWT2xC/y5JDctPLBIGzHKbhw==",
"version": "3.6.2",
"resolved": "https://registry.npmjs.org/prettier/-/prettier-3.6.2.tgz",
"integrity": "sha512-I7AIg5boAr5R0FFtJ6rCfD+LFsWHp81dolrFD8S79U9tb8Az2nGrJncnMSnys+bpQJfRUzqs9hnA81OAA3hCuQ==",
"dev": true,
"license": "MIT",
"bin": {
@ -13986,9 +14127,9 @@
}
},
"node_modules/undici-types": {
"version": "6.19.8",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-6.19.8.tgz",
"integrity": "sha512-ve2KP6f/JnbPBFyobGHuerC9g1FYGn/F8n1LWTwNxCEzd6IfqTwUQcNXgEtmmQ6DlRrC1hrSrBnCZPokRrDHjw==",
"version": "7.10.0",
"resolved": "https://registry.npmjs.org/undici-types/-/undici-types-7.10.0.tgz",
"integrity": "sha512-t5Fy/nfn+14LuOc2KNYg75vZqClpAiqscVvMygNnlsHBFpSXdJaYtXMcdNLpl/Qvc3P2cB3s6lOV51nqsFq4ag==",
"license": "MIT"
},
"node_modules/unified": {

12
llama_stack/ui/package.json
View file

@ -19,11 +19,11 @@
"@radix-ui/react-select": "^2.2.5",
"@radix-ui/react-separator": "^1.1.7",
"@radix-ui/react-slot": "^1.2.3",
"@radix-ui/react-tooltip": "^1.2.6",
"@radix-ui/react-tooltip": "^1.2.8",
"class-variance-authority": "^0.7.1",
"clsx": "^2.1.1",
"framer-motion": "^11.18.2",
"llama-stack-client": "^0.2.19",
"framer-motion": "^12.23.12",
"llama-stack-client": "^0.2.20",
"lucide-react": "^0.510.0",
"next": "15.3.3",
"next-auth": "^4.24.11",
@ -44,16 +44,16 @@
"@testing-library/jest-dom": "^6.8.0",
"@testing-library/react": "^16.3.0",
"@types/jest": "^29.5.14",
"@types/node": "^20",
"@types/node": "^24",
"@types/react": "^19",
"@types/react-dom": "^19",
"eslint": "^9",
"eslint-config-next": "15.3.2",
"eslint-config-next": "15.5.2",
"eslint-config-prettier": "^10.1.8",
"eslint-plugin-prettier": "^5.5.4",
"jest": "^29.7.0",
"jest-environment-jsdom": "^29.7.0",
"prettier": "3.5.3",
"prettier": "3.6.2",
"tailwindcss": "^4",
"ts-node": "^10.9.2",
"tw-animate-css": "^1.2.9",

19
pyproject.toml
View file

@ -7,7 +7,7 @@ required-version = ">=0.7.0"
[project]
name = "llama_stack"
version = "0.2.19"
version = "0.2.20"
authors = [{ name = "Meta Llama", email = "llama-oss@meta.com" }]
description = "Llama Stack"
readme = "README.md"
@ -31,9 +31,8 @@ dependencies = [
"huggingface-hub>=0.34.0,<1.0",
"jinja2>=3.1.6",
"jsonschema",
"llama-stack-client>=0.2.19",
"llama-api-client>=0.1.2",
"openai>=1.99.6,<1.100.0",
"llama-stack-client>=0.2.20",
"openai>=1.99.6",
"prompt-toolkit",
"python-dotenv",
"python-jose[cryptography]",
@ -56,7 +55,7 @@ dependencies = [
ui = [
"streamlit",
"pandas",
"llama-stack-client>=0.2.19",
"llama-stack-client>=0.2.20",
"streamlit-option-menu",
]
@ -84,6 +83,7 @@ unit = [
"openai",
"aiosqlite",
"aiohttp",
"psycopg2-binary>=2.9.0",
"pypdf",
"mcp",
"chardet",
@ -92,7 +92,7 @@ unit = [
"sqlalchemy[asyncio]>=2.0.41",
"blobfile",
"faiss-cpu",
"pymilvus>=2.5.12",
"pymilvus>=2.6.1",
"milvus-lite>=2.5.0",
"litellm",
"together",
@ -105,12 +105,13 @@ unit = [
# separately. If you are using "uv" to execute your tests, you can use the "--group" flag to specify extra
# dependencies.
test = [
"openai",
"openai>=1.100.0", # for expires_after support
"aiosqlite",
"aiohttp",
"torch>=2.6.0",
"torchvision>=0.21.0",
"chardet",
"psycopg2-binary>=2.9.0",
"pypdf",
"mcp",
"datasets",
@ -119,7 +120,7 @@ test = [
"sqlalchemy",
"sqlalchemy[asyncio]>=2.0.41",
"requests",
"pymilvus>=2.5.12",
"pymilvus>=2.6.1",
"milvus-lite>=2.5.0",
"weaviate-client>=4.16.4",
]
@ -144,7 +145,7 @@ docs = [
]
codegen = ["rich", "pydantic", "jinja2>=3.1.6"]
benchmark = [
"locust>=2.37.14",
"locust>=2.39.1",
]
[project.urls]

14
scripts/integration-tests.sh
View file

@ -140,13 +140,6 @@ THIS_DIR=$(dirname "$0")
ROOT_DIR="$THIS_DIR/.."
cd $ROOT_DIR
# Set recording directory
if [[ "$RUN_VISION_TESTS" == "true" ]]; then
export LLAMA_STACK_TEST_RECORDING_DIR="tests/integration/recordings/vision"
else
export LLAMA_STACK_TEST_RECORDING_DIR="tests/integration/recordings"
fi
# check if "llama" and "pytest" are available. this script does not use `uv run` given
# it can be used in a pre-release environment where we have not been able to tell
# uv about pre-release dependencies properly (yet).
@ -298,5 +291,12 @@ echo "=== System Resources After Tests ==="
free -h 2>/dev/null || echo "free command not available"
df -h
# stop server
if [[ "$STACK_CONFIG" == *"server:"* ]]; then
echo "Stopping Llama Stack Server..."
kill $(lsof -i :8321 | awk 'NR>1 {print $2}')
echo "Llama Stack Server stopped"
fi
echo ""
echo "=== Integration Tests Complete ==="

12
tests/README.md
View file

@ -38,26 +38,15 @@ For running integration tests, you must provide a few things:
- a distribution name (e.g., `starter`) or a path to a `run.yaml` file
- a comma-separated list of api=provider pairs, e.g. `inference=fireworks,safety=llama-guard,agents=meta-reference`. This is most useful for testing a single API surface.
- Whether you are using replay or live mode for inference. This is specified with the LLAMA_STACK_TEST_INFERENCE_MODE environment variable. The default mode currently is "live" -- that is certainly surprising, but we will fix this soon.
- Any API keys you need to use should be set in the environment, or can be passed in with the --env option.
You can run the integration tests in replay mode with:
```bash
# Run all tests with existing recordings
LLAMA_STACK_TEST_INFERENCE_MODE=replay \
LLAMA_STACK_TEST_RECORDING_DIR=tests/integration/recordings \
uv run --group test \
pytest -sv tests/integration/ --stack-config=starter
```
If you don't specify LLAMA_STACK_TEST_INFERENCE_MODE, by default it will be in "live" mode -- that is, it will make real API calls.
```bash
# Test against live APIs
FIREWORKS_API_KEY=your_key pytest -sv tests/integration/inference --stack-config=starter
```
### Re-recording tests
#### Local Re-recording (Manual Setup Required)
@ -66,7 +55,6 @@ If you want to re-record tests locally, you can do so with:
```bash
LLAMA_STACK_TEST_INFERENCE_MODE=record \
LLAMA_STACK_TEST_RECORDING_DIR=tests/integration/recordings \
uv run --group test \
pytest -sv tests/integration/ --stack-config=starter -k "<appropriate test name>"
```

19
tests/integration/README.md
View file

@ -98,29 +98,25 @@ pytest -s -v tests/integration/vector_io/ \
The testing system supports three modes controlled by environment variables:
### LIVE Mode (Default)
Tests make real API calls:
### REPLAY Mode (Default)
Uses cached responses instead of making API calls:
```bash
LLAMA_STACK_TEST_INFERENCE_MODE=live pytest tests/integration/
pytest tests/integration/
```
### RECORD Mode
Captures API interactions for later replay:
```bash
LLAMA_STACK_TEST_INFERENCE_MODE=record \
LLAMA_STACK_TEST_RECORDING_DIR=tests/integration/recordings \
pytest tests/integration/inference/test_new_feature.py
```
### REPLAY Mode
Uses cached responses instead of making API calls:
### LIVE Mode
Tests make real API calls (but not recorded):
```bash
LLAMA_STACK_TEST_INFERENCE_MODE=replay \
LLAMA_STACK_TEST_RECORDING_DIR=tests/integration/recordings \
pytest tests/integration/
LLAMA_STACK_TEST_INFERENCE_MODE=live pytest tests/integration/
```
Note that right now you must specify the recording directory. This is because different tests use different recording directories and we don't (yet) have a fool-proof way to map a test to a recording directory. We are working on this.
By default, the recording directory is `tests/integration/recordings`. You can override this by setting the `LLAMA_STACK_TEST_RECORDING_DIR` environment variable.
## Managing Recordings
@ -146,7 +142,6 @@ See the [main testing guide](../README.md#remote-re-recording-recommended) for f
```bash
# Re-record specific tests
LLAMA_STACK_TEST_INFERENCE_MODE=record \
LLAMA_STACK_TEST_RECORDING_DIR=tests/integration/recordings \
pytest -s -v --stack-config=server:starter tests/integration/inference/test_modified.py
```

2
tests/integration/conftest.py
View file

@ -30,6 +30,8 @@ def pytest_runtest_makereport(item, call):
def pytest_sessionstart(session):
# stop macOS from complaining about duplicate OpenMP libraries
os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"
if "LLAMA_STACK_TEST_INFERENCE_MODE" not in os.environ:
os.environ["LLAMA_STACK_TEST_INFERENCE_MODE"] = "replay"
def pytest_runtest_teardown(item):

83
tests/integration/files/test_files.py
View file

@ -8,6 +8,7 @@ from io import BytesIO
from unittest.mock import patch
import pytest
import requests
from llama_stack.core.datatypes import User
@ -79,6 +80,88 @@ def test_openai_client_basic_operations(openai_client):
pass # ignore 404
@pytest.mark.xfail(message="expires_after not available on all providers")
def test_expires_after(openai_client):
"""Test uploading a file with expires_after parameter."""
client = openai_client
uploaded_file = None
try:
with BytesIO(b"expires_after test") as file_buffer:
file_buffer.name = "expires_after.txt"
uploaded_file = client.files.create(
file=file_buffer,
purpose="assistants",
expires_after={"anchor": "created_at", "seconds": 4545},
)
assert uploaded_file.expires_at is not None
assert uploaded_file.expires_at == uploaded_file.created_at + 4545
listed = client.files.list()
ids = [f.id for f in listed.data]
assert uploaded_file.id in ids
retrieved = client.files.retrieve(uploaded_file.id)
assert retrieved.id == uploaded_file.id
finally:
if uploaded_file is not None:
try:
client.files.delete(uploaded_file.id)
except Exception:
pass
@pytest.mark.xfail(message="expires_after not available on all providers")
def test_expires_after_requests(openai_client):
"""Upload a file using requests multipart/form-data and bracketed expires_after fields.
This ensures clients that send form fields like `expires_after[anchor]` and
`expires_after[seconds]` are handled by the server.
"""
base_url = f"{openai_client.base_url}files"
uploaded_id = None
try:
files = {"file": ("expires_after_with_requests.txt", BytesIO(b"expires_after via requests"))}
data = {
"purpose": "assistants",
"expires_after[anchor]": "created_at",
"expires_after[seconds]": "4545",
}
session = requests.Session()
request = requests.Request("POST", base_url, files=files, data=data)
prepared = session.prepare_request(request)
resp = session.send(prepared, timeout=30)
resp.raise_for_status()
result = resp.json()
assert result.get("id", "").startswith("file-")
uploaded_id = result["id"]
assert result.get("created_at") is not None
assert result.get("expires_at") == result["created_at"] + 4545
list_resp = requests.get(base_url, timeout=30)
list_resp.raise_for_status()
listed = list_resp.json()
ids = [f["id"] for f in listed.get("data", [])]
assert uploaded_id in ids
retrieve_resp = requests.get(f"{base_url}/{uploaded_id}", timeout=30)
retrieve_resp.raise_for_status()
retrieved = retrieve_resp.json()
assert retrieved["id"] == uploaded_id
finally:
if uploaded_id:
try:
requests.delete(f"{base_url}/{uploaded_id}", timeout=30)
except Exception:
pass
@pytest.mark.xfail(message="User isolation broken for current providers, must be fixed.")
@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
def test_files_authentication_isolation(mock_get_authenticated_user, llama_stack_client):

10
tests/integration/recordings/responses/00ba04f74a96.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:53.860911Z",
"created_at": "2025-09-03T17:37:35.23084Z",
"done": true,
"done_reason": "stop",
"total_duration": 249137667,
"load_duration": 152509542,
"total_duration": 195981375,
"load_duration": 110522917,
"prompt_eval_count": 216,
"prompt_eval_duration": 71000000,
"prompt_eval_duration": 72393958,
"eval_count": 2,
"eval_duration": 24000000,
"eval_duration": 11843000,
"response": "safe",
"thinking": null,
"context": null

44
tests/integration/recordings/responses/04172112ffbb.json
View file

@ -21,7 +21,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.033900164Z",
"created_at": "2025-09-03T17:41:43.950283Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -39,7 +39,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.213371151Z",
"created_at": "2025-09-03T17:41:43.991122Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -57,7 +57,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.387513976Z",
"created_at": "2025-09-03T17:41:44.031378Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -75,7 +75,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.564344287Z",
"created_at": "2025-09-03T17:41:44.073098Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -93,7 +93,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.746579415Z",
"created_at": "2025-09-03T17:41:44.115961Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -111,7 +111,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:18.923276047Z",
"created_at": "2025-09-03T17:41:44.156517Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -129,7 +129,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.099961963Z",
"created_at": "2025-09-03T17:41:44.197079Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -147,7 +147,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.275621884Z",
"created_at": "2025-09-03T17:41:44.237565Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -165,7 +165,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.452204196Z",
"created_at": "2025-09-03T17:41:44.277755Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -183,7 +183,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.626937514Z",
"created_at": "2025-09-03T17:41:44.318476Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -201,7 +201,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.805566767Z",
"created_at": "2025-09-03T17:41:44.358628Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -219,7 +219,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:19.985987477Z",
"created_at": "2025-09-03T17:41:44.398984Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -237,7 +237,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:20.166458601Z",
"created_at": "2025-09-03T17:41:44.439232Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -255,7 +255,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:20.343346795Z",
"created_at": "2025-09-03T17:41:44.479478Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -273,7 +273,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:20.525008091Z",
"created_at": "2025-09-03T17:41:44.520202Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -291,7 +291,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:20.709087695Z",
"created_at": "2025-09-03T17:41:44.560517Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -309,7 +309,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:20.887074305Z",
"created_at": "2025-09-03T17:41:44.601592Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -327,15 +327,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:21.065244925Z",
"created_at": "2025-09-03T17:41:44.642064Z",
"done": true,
"done_reason": "stop",
"total_duration": 4373531496,
"load_duration": 44438132,
"total_duration": 887142667,
"load_duration": 119331417,
"prompt_eval_count": 56,
"prompt_eval_duration": 1296273199,
"prompt_eval_duration": 74294709,
"eval_count": 18,
"eval_duration": 3032321735,
"eval_duration": 692842791,
"response": "",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/0b27fd737699.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:13:57.556416Z",
"created_at": "2025-09-03T17:37:47.461886Z",
"done": true,
"done_reason": "stop",
"total_duration": 432363250,
"load_duration": 159296417,
"total_duration": 338927833,
"load_duration": 100895125,
"prompt_eval_count": 223,
"prompt_eval_duration": 257000000,
"prompt_eval_duration": 221583042,
"eval_count": 2,
"eval_duration": 14000000,
"eval_duration": 12341416,
"response": "safe",
"thinking": null,
"context": null

32
tests/integration/recordings/responses/0b3f2e4754ff.json
View file

@ -24,7 +24,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -39,7 +39,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921333,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -50,7 +50,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -65,7 +65,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921333,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -76,7 +76,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -91,7 +91,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921333,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -102,7 +102,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -117,7 +117,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921333,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -128,7 +128,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -143,7 +143,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921334,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -154,7 +154,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -169,7 +169,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921334,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -180,7 +180,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -195,7 +195,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921334,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -206,7 +206,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-29",
"id": "chatcmpl-414",
"choices": [
{
"delta": {
@ -221,7 +221,7 @@
"logprobs": null
}
],
"created": 1754090031,
"created": 1756921334,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,

12
tests/integration/recordings/responses/0e8f2b001dd9.json
View file

@ -1,7 +1,7 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/v1/v1/chat/completions",
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
"headers": {},
"body": {
"model": "llama3.2:3b-instruct-fp16",
@ -20,14 +20,14 @@
"body": {
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
"__data__": {
"id": "chatcmpl-368",
"id": "chatcmpl-161",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "Saturn is known for its extensive ring system.",
"content": "The answer is Saturn.",
"refusal": null,
"role": "assistant",
"annotations": null,
@ -37,15 +37,15 @@
}
}
],
"created": 1754081853,
"created": 1756921364,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion",
"service_tier": null,
"system_fingerprint": "fp_ollama",
"usage": {
"completion_tokens": 11,
"completion_tokens": 6,
"prompt_tokens": 39,
"total_tokens": 50,
"total_tokens": 45,
"completion_tokens_details": null,
"prompt_tokens_details": null
}

10
tests/integration/recordings/responses/10eea8c15ddc.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:51.682357Z",
"created_at": "2025-09-03T17:37:33.473237Z",
"done": true,
"done_reason": "stop",
"total_duration": 238161000,
"load_duration": 72494750,
"total_duration": 279025042,
"load_duration": 162673250,
"prompt_eval_count": 212,
"prompt_eval_duration": 87000000,
"prompt_eval_duration": 73595834,
"eval_count": 5,
"eval_duration": 74000000,
"eval_duration": 41950291,
"response": "unsafe\nS8",
"thinking": null,
"context": null

12
tests/integration/recordings/responses/140187e305dc.json
View file

@ -1,7 +1,7 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/v1/v1/chat/completions",
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
"headers": {},
"body": {
"model": "llama3.2:3b-instruct-fp16",
@ -20,14 +20,14 @@
"body": {
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
"__data__": {
"id": "chatcmpl-876",
"id": "chatcmpl-974",
"choices": [
{
"finish_reason": "stop",
"index": 0,
"logprobs": null,
"message": {
"content": "I'm afraid I don't have a built-in ability to directly interface with or \"test\" OpenAI models, including the original GPT-1 model. However, I can explain how you might approach this task:\n\nThe OpenAI GPT-1 is a large transformer-based language model that was trained on a massive dataset of text and achieved state-of-the-art results in various natural language processing tasks.\n\nTo test or evaluate the performance of a model like GPT-1, you would typically follow these steps:\n\n1. **Get access to the OpenAI API**: The OpenAI API provides a way for developers to interact with the GPT-1 model programmatically. You can sign up for an API key on the OpenAI website.\n2. **Choose a testing platform or environment**: You'll need a compute platform that supports the necessary algorithms and data structures to run inference on the GPT-1 model. Some popular options include AWS, Google Cloud, or Azure Compute Virtual Machines.\n3. **Prepare your test input data**: This will involve creating text inputs in the format expected by the OpenAI API (i.e., a JSON object containing the text to be processed).\n4. **Use the OpenAI Python library or SDK**: The OpenAI Python library provides an easy-to-use interface for interacting with the GPT-1 model through the API.\n\nHere's some example code that demonstrates how you might use the OpenAI Flask API to test a single input:\n\n```python\nfrom flask import Flask, request, jsonify\nimport json\n\napp = Flask(__name__)\n\n@ app . route ( '/ /gpt-en ', ' Text ', methods = ['POST'])\ndef gpt_en () -> Json :\n data = request . get_json ()\n if not data or \"message\" in ( data ):\n return None , 400 , { ' error' : \"Input must be a text string.\" }\n response = []\n while True:\n message = \"\"\n for token in data [\"input\"]:\n response_text = f\"{data['prompt']} {token}\"\n data[\"input\"] = [response_text]\n new_response = gpt_en()(data)\n if all([not item or not isinstance(item, dict) for item in new_response]):\n break\n\n message = json . dumps ({}\"text\": response_text})\n response.append(message)\n\n return jsonify ({\"output\": response}), 200 , {}\n\nif __name__ == \"__main__\":\n app.run(debug=True)\n```\n\n5. **Evaluate the output**: Once you have processed your test input data using the GPT-1 model, you can evaluate the accuracy of the generated responses.\n\nKeep in mind that this is just a basic example to illustrate how you might approach testing the OpenAI GPT-1 model.",
"content": "I'm happy to help you test the OpenAI API, however I can not access the API.\n\nInstead why don't we follow these steps:\n\n* Check documentation\n* Contact support\n* Reach out to their community forum. \n\nLet me know if I can be of any additional assistance",
"refusal": null,
"role": "assistant",
"annotations": null,
@ -37,15 +37,15 @@
}
}
],
"created": 1754510050,
"created": 1756921202,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion",
"service_tier": null,
"system_fingerprint": "fp_ollama",
"usage": {
"completion_tokens": 567,
"completion_tokens": 61,
"prompt_tokens": 31,
"total_tokens": 598,
"total_tokens": 92,
"completion_tokens_details": null,
"prompt_tokens_details": null
}

10
tests/integration/recordings/responses/17253d7cc667.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:52.919624Z",
"created_at": "2025-09-03T17:37:34.308033Z",
"done": true,
"done_reason": "stop",
"total_duration": 201956834,
"load_duration": 105132584,
"total_duration": 200296000,
"load_duration": 115974708,
"prompt_eval_count": 212,
"prompt_eval_duration": 75000000,
"prompt_eval_duration": 72173459,
"eval_count": 2,
"eval_duration": 20000000,
"eval_duration": 11536750,
"response": "safe",
"thinking": null,
"context": null

32
tests/integration/recordings/responses/173ecb3aab28.json
View file

@ -40,7 +40,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -55,7 +55,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -66,7 +66,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -81,7 +81,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -92,7 +92,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -107,7 +107,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -118,7 +118,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -133,7 +133,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -144,7 +144,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -159,7 +159,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -170,7 +170,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -185,7 +185,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -196,7 +196,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -211,7 +211,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -222,7 +222,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-457",
"id": "chatcmpl-921",
"choices": [
{
"delta": {
@ -237,7 +237,7 @@
"logprobs": null
}
],
"created": 1754090032,
"created": 1756920971,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,

10
tests/integration/recordings/responses/174458ad71b2.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:53.580806Z",
"created_at": "2025-09-03T17:37:34.994704Z",
"done": true,
"done_reason": "stop",
"total_duration": 205732750,
"load_duration": 98967000,
"total_duration": 339570875,
"load_duration": 262794125,
"prompt_eval_count": 213,
"prompt_eval_duration": 86000000,
"prompt_eval_duration": 64061000,
"eval_count": 2,
"eval_duration": 18000000,
"eval_duration": 11839042,
"response": "safe",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/178016edef0e.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:52.354566Z",
"created_at": "2025-09-03T17:37:33.769233Z",
"done": true,
"done_reason": "stop",
"total_duration": 605192500,
"load_duration": 457087166,
"total_duration": 253836584,
"load_duration": 138624959,
"prompt_eval_count": 210,
"prompt_eval_duration": 63000000,
"prompt_eval_duration": 69496125,
"eval_count": 5,
"eval_duration": 84000000,
"eval_duration": 45062833,
"response": "unsafe\nS12",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/197228e26971.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:52.686478Z",
"created_at": "2025-09-03T17:37:34.074233Z",
"done": true,
"done_reason": "stop",
"total_duration": 304136208,
"load_duration": 155977000,
"total_duration": 270746375,
"load_duration": 156423042,
"prompt_eval_count": 213,
"prompt_eval_duration": 71000000,
"prompt_eval_duration": 70338083,
"eval_count": 5,
"eval_duration": 76000000,
"eval_duration": 43379167,
"response": "unsafe\nS2",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/198ef7208389.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:51.186501Z",
"created_at": "2025-09-03T17:37:32.84197Z",
"done": true,
"done_reason": "stop",
"total_duration": 3146184459,
"load_duration": 2533467917,
"total_duration": 21572898667,
"load_duration": 21155275042,
"prompt_eval_count": 212,
"prompt_eval_duration": 526000000,
"prompt_eval_duration": 371898125,
"eval_count": 5,
"eval_duration": 83000000,
"eval_duration": 43290458,
"response": "unsafe\nS1",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/1adfaa0e062e.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:12:53.332041Z",
"created_at": "2025-09-03T17:37:34.607413Z",
"done": true,
"done_reason": "stop",
"total_duration": 365895333,
"load_duration": 257825208,
"total_duration": 267812042,
"load_duration": 181570000,
"prompt_eval_count": 213,
"prompt_eval_duration": 78000000,
"prompt_eval_duration": 73947375,
"eval_count": 2,
"eval_duration": 28000000,
"eval_duration": 11708000,
"response": "safe",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/1b8394f90636.json
View file

@ -22,15 +22,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:05.685988Z",
"created_at": "2025-09-03T17:36:13.821929Z",
"done": true,
"done_reason": "stop",
"total_duration": 14128980625,
"load_duration": 7220159208,
"total_duration": 1907912167,
"load_duration": 90979292,
"prompt_eval_count": 18,
"prompt_eval_duration": 4658000000,
"prompt_eval_duration": 77350291,
"eval_count": 43,
"eval_duration": 2224000000,
"eval_duration": 1738568334,
"response": " _______.\n\nThe best answer is blue. The traditional nursery rhyme goes like this:\n\nRoses are red,\nViolets are blue,\nSugar is sweet,\nAnd so are you! (Or something similar.)",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/1b92be674e2a.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:50:06.140190726Z",
"created_at": "2025-09-03T17:39:38.236797Z",
"done": true,
"done_reason": "stop",
"total_duration": 5213341378,
"load_duration": 43943569,
"total_duration": 1296281500,
"load_duration": 283393917,
"prompt_eval_count": 23,
"prompt_eval_duration": 1049424427,
"prompt_eval_duration": 75453042,
"eval_count": 24,
"eval_duration": 4119422888,
"eval_duration": 936860125,
"response": "Mark Zuckerberg is the founder, chairman and CEO of Meta, which he originally founded as Facebook in 2004.",
"thinking": null,
"context": null

422
tests/integration/recordings/responses/1e11c2b20ff8.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"How do systems learn automatically?"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
0.042460807,
-0.06189971,
-0.0784711,
0.0064329687,
0.03129365,
0.00807445,
0.05801836,
0.025447326,
0.016402787,
0.045995634,
-0.028924342,
0.04451832,
0.05686613,
-0.015340794,
-0.07020505,
-0.057178136,
-0.07683263,
0.006748679,
0.0043323045,
-0.123651944,
0.0031534543,
-0.03258051,
-0.02936216,
0.024140852,
-0.028559243,
0.10224467,
0.0021632623,
-0.006975691,
0.025292527,
-0.055500276,
0.031231727,
-0.0070274337,
0.08430815,
-0.028431177,
-0.083029,
0.009555893,
-0.020029299,
-0.00243229,
-0.00768719,
-0.023077851,
-0.09293533,
-0.042625993,
-0.020000124,
0.008240663,
0.060970567,
0.050315727,
-0.0510085,
-0.008543903,
-0.030227834,
-0.03582846,
-0.17836656,
-0.047279052,
0.033892106,
0.031623542,
-0.008832113,
0.10480918,
0.033559043,
0.090348184,
-0.015757555,
-0.0125672715,
-0.084686965,
-0.114781834,
-0.13755985,
0.021652374,
0.047834594,
0.043243896,
0.008659893,
0.038724966,
0.046716973,
-0.077413626,
-0.04887495,
0.031287406,
0.022356613,
0.00043283988,
0.052321073,
-0.012254071,
-0.035172574,
-0.00825216,
-0.008866574,
-0.034267236,
-0.04576201,
0.002467568,
-0.040877618,
0.08047682,
0.09472728,
0.0413438,
0.0057974122,
0.044982508,
0.025369909,
0.006618073,
0.010467276,
-0.07960384,
-0.03108485,
-0.03528749,
0.01831391,
0.053473305,
0.06568304,
-0.07259002,
0.02523736,
0.10520362,
0.035732146,
0.028157586,
0.011687256,
0.044207197,
0.012604437,
0.0018819098,
0.03926183,
0.043135095,
0.09784739,
-0.08801336,
-0.06060836,
0.02681984,
0.0041358666,
0.033492945,
0.011799116,
0.009551661,
-0.0095491735,
-0.021212189,
-0.008917248,
0.029352615,
-0.012693442,
-0.019269384,
0.009901157,
-0.00812101,
0.018603146,
-0.0007501193,
-0.056115113,
-3.8018077e-33,
0.020848714,
0.0047160466,
0.019726405,
0.06024251,
-0.0685974,
-0.07497267,
0.007997452,
-0.047339544,
0.057801835,
0.049544968,
0.01878086,
0.03274472,
0.017663997,
0.07483022,
0.02496901,
-0.011843339,
-0.11212756,
0.0070379525,
0.028099466,
-0.01746246,
0.08173482,
-0.007920462,
0.032095373,
-0.12300146,
0.033773854,
0.025873141,
-0.0045020077,
0.079493225,
0.0040725255,
0.03305898,
0.008061117,
0.0134422695,
-0.03292251,
0.031554114,
0.04013794,
0.0014983519,
0.030762345,
0.029481992,
0.041350223,
-0.047438618,
0.03944708,
-0.07526981,
0.037927423,
-0.026016014,
0.016933467,
0.0136799775,
0.0071263947,
-0.05386736,
-0.07443268,
-0.006070775,
0.024427462,
-0.039844982,
-0.020661902,
-0.033354662,
0.009005565,
0.12111172,
-0.028260944,
-0.036192853,
-0.021332363,
0.05333571,
0.05161245,
-0.01204843,
0.035563566,
0.05408247,
0.060722187,
0.07159865,
0.04299143,
0.008544481,
0.07421879,
0.00841512,
-0.036342908,
-0.008549791,
-0.08816386,
-0.049075164,
0.00029373015,
-0.05127952,
0.03586739,
-0.030380003,
-0.012642127,
0.018771531,
0.01711824,
-0.06644723,
0.023793438,
0.0010271219,
-0.01939443,
-0.053452212,
-0.017060323,
-0.062207118,
-0.05962535,
-0.012172617,
-0.013190802,
-0.037036054,
0.00082622556,
0.098088354,
0.024690514,
2.1767905e-33,
-0.010088812,
-0.016811697,
-0.042140447,
0.08837209,
-0.028899776,
-0.0048947735,
-0.082139015,
0.029238816,
-0.043079354,
-0.014153092,
-0.028387645,
0.025998218,
-0.017625,
0.046511114,
-0.005768211,
0.030010609,
0.011375536,
0.017426634,
0.055062976,
0.032230247,
-0.07995765,
0.032486655,
-0.060016844,
-0.011561194,
0.010211269,
0.046528235,
0.001191399,
0.0786961,
-0.0446158,
0.032789085,
0.0023115936,
-0.03886269,
-0.017663589,
0.07913024,
-0.004583343,
0.043521065,
-0.031589273,
0.008867868,
-0.05013296,
0.068929516,
0.043675046,
0.019968731,
-0.08471742,
-0.046864275,
-0.0068198936,
-0.026138468,
-0.05107216,
0.054374695,
0.03069186,
-0.010925094,
0.04721093,
-0.017387696,
-0.020754937,
-0.081763394,
-0.027709637,
0.035980806,
0.05396534,
0.044874854,
0.059699643,
0.041227758,
-0.06664364,
-0.09201654,
0.008915574,
0.025849758,
-0.038651932,
-0.0044070315,
-0.052066546,
0.027435115,
0.012089562,
0.048306923,
0.059854515,
0.097325735,
-0.053612895,
-0.07639326,
0.015773866,
-0.0444848,
-0.13214406,
-0.0702488,
-0.10134438,
-0.11905995,
-0.027714504,
0.006891868,
-0.0053650527,
0.054135524,
-0.111159205,
0.07835098,
0.03506018,
0.016036613,
0.021490784,
-0.061526407,
0.007425222,
0.04833579,
-0.01361202,
0.012450488,
-0.12729599,
-1.4009424e-08,
-0.040908325,
-0.01596458,
0.060048707,
0.03804525,
0.0663794,
0.04727275,
-0.016112225,
0.09687414,
-0.04424251,
-0.028799534,
-0.01294642,
0.013026413,
0.022404836,
0.04713173,
0.06402557,
0.12130648,
0.06062839,
0.10218965,
-0.0757528,
-0.023806982,
0.12489501,
-0.045460615,
0.09545599,
0.021262301,
0.03731495,
-0.075220875,
-0.0026194793,
0.0472452,
0.048499025,
0.12358729,
0.017998053,
0.013811017,
-0.035893846,
-0.051789004,
0.06182457,
0.05160056,
0.008895317,
-0.12500942,
0.016453298,
-0.08590811,
-0.071096726,
0.06987216,
-0.036072273,
-0.0053715096,
-0.048762616,
0.00081640907,
-0.021502526,
-0.061078615,
0.002485032,
-0.032720752,
0.045743283,
0.038934175,
-0.024666062,
0.025897244,
0.10301431,
-0.013001504,
0.04783332,
-0.07114252,
0.046031926,
0.080549754,
-0.10302451,
0.08449227,
0.028010191,
-0.03697792
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 6,
"total_tokens": 6
}
}
},
"is_streaming": false
}
}

10
tests/integration/recordings/responses/211b1562d4e6.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.15982Z",
"created_at": "2025-09-03T17:36:17.894986Z",
"done": true,
"done_reason": "stop",
"total_duration": 498612042,
"load_duration": 71411834,
"total_duration": 363397458,
"load_duration": 86692791,
"prompt_eval_count": 23,
"prompt_eval_duration": 102000000,
"prompt_eval_duration": 68658541,
"eval_count": 6,
"eval_duration": 323000000,
"eval_duration": 207389084,
"response": "Humans live on Earth.",
"thinking": null,
"context": null

422
tests/integration/recordings/responses/23506e73bb9e.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"This is a test file 1"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.055990793,
0.076004684,
-0.09247725,
0.014340361,
0.058780864,
-0.032434482,
0.020954052,
0.028818125,
-0.06591213,
0.013541593,
0.12999941,
0.004603084,
-0.0069239275,
-0.055457443,
-0.047553156,
-0.029139794,
-0.12236376,
-0.05360872,
-0.014706594,
0.05984688,
0.034442738,
0.02076038,
-0.048697792,
0.0135388365,
0.058592733,
-0.003076384,
-0.031565297,
0.082541116,
-0.031259205,
-0.12057633,
0.038319625,
0.06574785,
0.06415721,
0.038382582,
0.12570712,
0.03108174,
0.10821103,
-0.0019794356,
-0.024704305,
0.028765837,
0.01268161,
-0.039844505,
0.043253522,
-0.015898596,
-0.0135526005,
-0.0050831717,
-0.007911988,
0.039783813,
0.0036548872,
-0.033632487,
-0.058547974,
0.0048877494,
-0.089586094,
-0.010457663,
0.059202507,
-0.020414542,
0.014278556,
0.013986488,
-0.0046022516,
0.0383391,
0.0048145773,
0.029772853,
-0.020863408,
0.018640704,
0.12422993,
-0.023236223,
-0.040323637,
-0.023598222,
-0.007448043,
-0.09083128,
-0.16859712,
0.01012451,
-0.035808884,
0.010595173,
-0.02050494,
0.0020821376,
-0.10925222,
0.00793264,
0.048889533,
-0.11391199,
-0.06072707,
-0.13435508,
0.0063265716,
-0.008838073,
-0.03153269,
0.099169336,
0.055310693,
0.0068571265,
-0.023463152,
-0.0031599961,
0.036782328,
0.014336826,
0.022220163,
0.047114056,
0.007079763,
0.06806425,
0.01851431,
0.040882625,
0.055058856,
0.09488346,
-0.015833577,
-7.924328e-05,
0.010821554,
0.09177704,
-0.07464829,
-0.06471165,
0.07013805,
-0.04499751,
0.057702336,
-0.0260911,
0.006323043,
-0.09500501,
-0.010549514,
-0.07887475,
0.039744847,
-0.04154404,
-0.055268157,
0.07540271,
-0.04667509,
0.036143072,
0.080297194,
-0.036381353,
-0.03477274,
0.01701203,
-0.047007203,
-0.06519774,
0.062141683,
-4.222482e-33,
-0.0017580023,
-0.09383388,
-0.02982657,
0.1257841,
0.03802007,
-0.03654342,
0.0060920226,
0.05906885,
-0.11074452,
0.005664566,
-0.0259852,
-0.074819505,
0.008342821,
0.027451068,
-0.05248069,
0.02401768,
-0.004380289,
0.039321493,
-0.04213744,
-0.027290314,
0.054677974,
0.02707243,
-0.03329442,
-0.060589895,
-0.050737355,
0.017969057,
-0.0035060972,
-0.04666249,
0.073946096,
0.01333894,
-0.0033873583,
-0.046544433,
-0.060105033,
0.03406923,
0.001542676,
0.039177947,
0.03989323,
-0.012346489,
-0.030511485,
-0.0019157606,
-0.014608986,
-0.012997742,
0.019522104,
-0.022349002,
0.074362256,
-0.053366993,
-0.023993475,
0.029225096,
0.027534606,
0.015111057,
-0.020442221,
0.043327376,
0.019660354,
0.017330697,
-0.0035011724,
0.019482937,
-0.0003428041,
0.0004143988,
-0.005117252,
0.06624799,
0.027922852,
0.041020587,
-0.067166425,
0.028737254,
-0.03478325,
-0.055551115,
-0.032713737,
-0.08099247,
0.09216284,
0.06395264,
-0.049168136,
-0.039908994,
0.036915958,
-0.001602359,
0.00033041168,
-0.026015632,
-0.005999889,
0.05474541,
-0.09568287,
-0.05186289,
-0.048838183,
-0.08639551,
-0.034023147,
-0.033257127,
-0.05651867,
-0.051131375,
0.00809173,
-0.08581851,
0.06507323,
-0.085427366,
0.027997404,
0.029847065,
-0.031673994,
-0.08560956,
0.1017672,
2.1855676e-33,
0.01160785,
0.077607885,
-0.017380483,
0.005239329,
0.0009684126,
0.06543702,
0.07256893,
-0.044318836,
-0.04749324,
0.14031002,
-0.025741624,
0.0057860985,
0.040946104,
-0.054880083,
0.074413285,
-0.023610368,
0.018364722,
-0.060585637,
-0.044149306,
0.0027854694,
-0.04580664,
0.1172219,
0.10268574,
0.07907412,
-0.0466143,
0.018618405,
0.029834948,
0.037265483,
0.02273822,
-0.0026589038,
0.041726097,
0.06439532,
-0.089163445,
0.018188318,
0.024064727,
-0.096389584,
0.08642254,
-0.05389359,
0.01923105,
0.045092683,
0.045125954,
0.09655961,
0.014908797,
0.059611585,
0.03066662,
0.05882299,
0.111484826,
0.016632542,
0.011590394,
-0.023702666,
-0.008617484,
-0.055030316,
0.047606383,
-0.014632687,
-0.014156344,
0.069926,
0.032047603,
0.042642817,
-0.053942375,
0.031047028,
0.009216673,
0.033024028,
-0.019033706,
0.005568194,
-0.014985451,
-0.09193244,
-0.03210824,
0.015367608,
0.029150328,
0.01250386,
-0.004827391,
0.023345906,
-0.028271332,
-0.08454125,
0.051068563,
-0.0133641455,
-0.029022738,
-0.02258452,
0.010884119,
-0.009810021,
0.049751773,
-0.0032637494,
-0.038813565,
0.027924104,
0.017925078,
0.005337612,
0.058691237,
0.09577674,
-0.014308608,
0.006972794,
-0.02733344,
0.06912433,
0.05727631,
0.03206042,
0.0042422824,
-1.6766318e-08,
-0.036354303,
-0.09146416,
-0.026319364,
-0.007941995,
-0.024127059,
0.09896698,
-0.04723083,
-0.03767135,
-0.029419973,
-0.022513283,
0.04125822,
-0.0011487947,
-0.05570366,
0.020679709,
-0.038118906,
-0.0524994,
-0.02624128,
-0.05336954,
-0.040593866,
-0.0073642326,
-0.0014442836,
0.02714257,
0.027141048,
0.00932513,
-0.00026505854,
0.038233075,
0.037096914,
0.08405413,
-0.06340637,
-0.014856458,
0.05038612,
0.06703033,
0.027668556,
-0.04360097,
-0.012041474,
0.08500689,
0.111594744,
0.1046117,
0.019726463,
-0.0003025109,
-0.04110389,
0.009575226,
-0.05285304,
-0.0026365265,
-0.031144748,
-0.08860188,
-0.06762232,
-0.07451522,
-0.053012833,
-0.09560941,
-0.05273455,
0.013032144,
0.0029190276,
0.041905046,
-0.04522114,
0.016730292,
0.017214278,
0.021578068,
-0.03718778,
0.02353425,
0.052041385,
0.06444499,
0.02387539,
-0.025236009
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 6,
"total_tokens": 6
}
}
},
"is_streaming": false
}
}

34
tests/integration/recordings/responses/2afe3b38ca01.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:01.887809Z",
"created_at": "2025-09-03T17:37:50.436472Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:01.942369Z",
"created_at": "2025-09-03T17:37:50.478138Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:01.99605Z",
"created_at": "2025-09-03T17:37:50.519952Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.049974Z",
"created_at": "2025-09-03T17:37:50.561433Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.102027Z",
"created_at": "2025-09-03T17:37:50.603624Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.158416Z",
"created_at": "2025-09-03T17:37:50.645851Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.211753Z",
"created_at": "2025-09-03T17:37:50.688403Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.265564Z",
"created_at": "2025-09-03T17:37:50.72991Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.31618Z",
"created_at": "2025-09-03T17:37:50.771635Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.370325Z",
"created_at": "2025-09-03T17:37:50.813711Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.424667Z",
"created_at": "2025-09-03T17:37:50.856201Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.47913Z",
"created_at": "2025-09-03T17:37:50.899048Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,15 +238,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:02.536984Z",
"created_at": "2025-09-03T17:37:50.94069Z",
"done": true,
"done_reason": "stop",
"total_duration": 1042724125,
"load_duration": 86161375,
"total_duration": 688370708,
"load_duration": 107469833,
"prompt_eval_count": 399,
"prompt_eval_duration": 305000000,
"prompt_eval_duration": 74988334,
"eval_count": 13,
"eval_duration": 650000000,
"eval_duration": 505216458,
"response": "",
"thinking": null,
"context": null

208
tests/integration/recordings/responses/2d187a11704c.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:11.938867Z",
"created_at": "2025-09-03T17:37:56.566151Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:11.991247Z",
"created_at": "2025-09-03T17:37:56.609308Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.043953Z",
"created_at": "2025-09-03T17:37:56.651314Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.096001Z",
"created_at": "2025-09-03T17:37:56.693185Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.150454Z",
"created_at": "2025-09-03T17:37:56.734643Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.201249Z",
"created_at": "2025-09-03T17:37:56.776343Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.252534Z",
"created_at": "2025-09-03T17:37:56.81705Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.30063Z",
"created_at": "2025-09-03T17:37:56.857959Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.351034Z",
"created_at": "2025-09-03T17:37:56.899424Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.405032Z",
"created_at": "2025-09-03T17:37:56.939218Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.462645Z",
"created_at": "2025-09-03T17:37:56.980065Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.520337Z",
"created_at": "2025-09-03T17:37:57.02214Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,7 +238,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.575809Z",
"created_at": "2025-09-03T17:37:57.0628Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -256,7 +256,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.633724Z",
"created_at": "2025-09-03T17:37:57.106061Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -274,7 +274,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.683133Z",
"created_at": "2025-09-03T17:37:57.1492Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -292,7 +292,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.734309Z",
"created_at": "2025-09-03T17:37:57.190075Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -310,7 +310,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.785917Z",
"created_at": "2025-09-03T17:37:57.23178Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -328,7 +328,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.835705Z",
"created_at": "2025-09-03T17:37:57.272738Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -346,7 +346,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.886509Z",
"created_at": "2025-09-03T17:37:57.313855Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -364,7 +364,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.937134Z",
"created_at": "2025-09-03T17:37:57.354964Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -382,7 +382,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:12.988532Z",
"created_at": "2025-09-03T17:37:57.395971Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -400,7 +400,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.041798Z",
"created_at": "2025-09-03T17:37:57.438471Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -418,7 +418,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.095443Z",
"created_at": "2025-09-03T17:37:57.479796Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -436,7 +436,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.151402Z",
"created_at": "2025-09-03T17:37:57.520641Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -454,7 +454,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.203462Z",
"created_at": "2025-09-03T17:37:57.561511Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -472,7 +472,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.254567Z",
"created_at": "2025-09-03T17:37:57.602875Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -490,7 +490,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.305865Z",
"created_at": "2025-09-03T17:37:57.643406Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -508,7 +508,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.357658Z",
"created_at": "2025-09-03T17:37:57.684279Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -526,7 +526,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.407773Z",
"created_at": "2025-09-03T17:37:57.725699Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -544,7 +544,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.458919Z",
"created_at": "2025-09-03T17:37:57.766658Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -562,7 +562,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.510456Z",
"created_at": "2025-09-03T17:37:57.80738Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -580,7 +580,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.565948Z",
"created_at": "2025-09-03T17:37:57.848466Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -598,7 +598,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.619155Z",
"created_at": "2025-09-03T17:37:57.889056Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -616,7 +616,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.672754Z",
"created_at": "2025-09-03T17:37:57.931554Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -634,7 +634,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.729473Z",
"created_at": "2025-09-03T17:37:57.974754Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -652,7 +652,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.788666Z",
"created_at": "2025-09-03T17:37:58.016978Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -670,7 +670,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.850575Z",
"created_at": "2025-09-03T17:37:58.057942Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -688,7 +688,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.904807Z",
"created_at": "2025-09-03T17:37:58.099015Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -706,7 +706,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:13.958524Z",
"created_at": "2025-09-03T17:37:58.140531Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -724,7 +724,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.011742Z",
"created_at": "2025-09-03T17:37:58.181382Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -742,7 +742,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.064933Z",
"created_at": "2025-09-03T17:37:58.223318Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -760,7 +760,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.116454Z",
"created_at": "2025-09-03T17:37:58.26358Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -778,7 +778,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.172682Z",
"created_at": "2025-09-03T17:37:58.305496Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -796,7 +796,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.227654Z",
"created_at": "2025-09-03T17:37:58.347254Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -814,7 +814,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.282068Z",
"created_at": "2025-09-03T17:37:58.390044Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -832,7 +832,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.334565Z",
"created_at": "2025-09-03T17:37:58.430867Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -850,7 +850,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.383532Z",
"created_at": "2025-09-03T17:37:58.471376Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -868,7 +868,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.432138Z",
"created_at": "2025-09-03T17:37:58.51208Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -886,7 +886,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.480995Z",
"created_at": "2025-09-03T17:37:58.553226Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -904,7 +904,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.531968Z",
"created_at": "2025-09-03T17:37:58.594787Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -922,7 +922,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.584044Z",
"created_at": "2025-09-03T17:37:58.63466Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -940,7 +940,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.635691Z",
"created_at": "2025-09-03T17:37:58.674628Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -958,7 +958,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.68837Z",
"created_at": "2025-09-03T17:37:58.714616Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -976,7 +976,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.73985Z",
"created_at": "2025-09-03T17:37:58.754906Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -994,7 +994,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.792412Z",
"created_at": "2025-09-03T17:37:58.795048Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1012,7 +1012,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.845872Z",
"created_at": "2025-09-03T17:37:58.835297Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1030,7 +1030,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.900102Z",
"created_at": "2025-09-03T17:37:58.875738Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1048,7 +1048,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:14.954589Z",
"created_at": "2025-09-03T17:37:58.91604Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1066,7 +1066,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.006629Z",
"created_at": "2025-09-03T17:37:58.956596Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1084,7 +1084,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.058561Z",
"created_at": "2025-09-03T17:37:58.996664Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1102,7 +1102,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.111954Z",
"created_at": "2025-09-03T17:37:59.037796Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1120,7 +1120,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.169173Z",
"created_at": "2025-09-03T17:37:59.078586Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1138,7 +1138,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.222569Z",
"created_at": "2025-09-03T17:37:59.119448Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1156,7 +1156,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.275795Z",
"created_at": "2025-09-03T17:37:59.160318Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1174,7 +1174,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.3327Z",
"created_at": "2025-09-03T17:37:59.201852Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1192,7 +1192,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.389931Z",
"created_at": "2025-09-03T17:37:59.243763Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1210,7 +1210,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.442349Z",
"created_at": "2025-09-03T17:37:59.284948Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1228,7 +1228,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.494175Z",
"created_at": "2025-09-03T17:37:59.325598Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1246,7 +1246,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.545764Z",
"created_at": "2025-09-03T17:37:59.366289Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1264,7 +1264,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.599099Z",
"created_at": "2025-09-03T17:37:59.406764Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1282,7 +1282,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.649852Z",
"created_at": "2025-09-03T17:37:59.447922Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1300,7 +1300,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.698222Z",
"created_at": "2025-09-03T17:37:59.488486Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1318,7 +1318,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.747168Z",
"created_at": "2025-09-03T17:37:59.529Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1336,7 +1336,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.797196Z",
"created_at": "2025-09-03T17:37:59.569417Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1354,7 +1354,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.845587Z",
"created_at": "2025-09-03T17:37:59.610542Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1372,7 +1372,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.897171Z",
"created_at": "2025-09-03T17:37:59.651411Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1390,7 +1390,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.944524Z",
"created_at": "2025-09-03T17:37:59.69241Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1408,7 +1408,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:15.994467Z",
"created_at": "2025-09-03T17:37:59.732339Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1426,7 +1426,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.045224Z",
"created_at": "2025-09-03T17:37:59.772462Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1444,7 +1444,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.093853Z",
"created_at": "2025-09-03T17:37:59.812507Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1462,7 +1462,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.144847Z",
"created_at": "2025-09-03T17:37:59.852762Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1480,7 +1480,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.197888Z",
"created_at": "2025-09-03T17:37:59.892984Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1498,7 +1498,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.250854Z",
"created_at": "2025-09-03T17:37:59.933555Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1516,7 +1516,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.301995Z",
"created_at": "2025-09-03T17:37:59.973778Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1534,7 +1534,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.352508Z",
"created_at": "2025-09-03T17:38:00.014923Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1552,7 +1552,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.40259Z",
"created_at": "2025-09-03T17:38:00.057464Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1570,7 +1570,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.453514Z",
"created_at": "2025-09-03T17:38:00.09902Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1588,7 +1588,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.50378Z",
"created_at": "2025-09-03T17:38:00.140492Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1606,7 +1606,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.554395Z",
"created_at": "2025-09-03T17:38:00.180239Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1624,7 +1624,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.605795Z",
"created_at": "2025-09-03T17:38:00.220364Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1642,7 +1642,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.656313Z",
"created_at": "2025-09-03T17:38:00.26097Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1660,7 +1660,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.706438Z",
"created_at": "2025-09-03T17:38:00.301228Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1678,7 +1678,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.756444Z",
"created_at": "2025-09-03T17:38:00.341631Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1696,7 +1696,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.807687Z",
"created_at": "2025-09-03T17:38:00.383006Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1714,7 +1714,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.85835Z",
"created_at": "2025-09-03T17:38:00.423509Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1732,7 +1732,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.909311Z",
"created_at": "2025-09-03T17:38:00.464702Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1750,7 +1750,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:16.959327Z",
"created_at": "2025-09-03T17:38:00.505914Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1768,7 +1768,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:17.010211Z",
"created_at": "2025-09-03T17:38:00.546505Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1786,7 +1786,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:17.061365Z",
"created_at": "2025-09-03T17:38:00.587839Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -1804,15 +1804,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:17.111956Z",
"created_at": "2025-09-03T17:38:00.629018Z",
"done": true,
"done_reason": "stop",
"total_duration": 5499672375,
"load_duration": 58161750,
"total_duration": 4303339291,
"load_duration": 156231250,
"prompt_eval_count": 36,
"prompt_eval_duration": 266000000,
"prompt_eval_duration": 81909875,
"eval_count": 100,
"eval_duration": 5174000000,
"eval_duration": 4064559292,
"response": "",
"thinking": null,
"context": null

758
tests/integration/recordings/responses/325a72db5755.json
View file

File diff suppressed because it is too large Load diff

8
tests/integration/recordings/responses/382c2f22274c.json
View file

@ -1,7 +1,7 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/v1/v1/chat/completions",
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
"headers": {},
"body": {
"model": "llama3.2:3b-instruct-fp16",
@ -22,14 +22,14 @@
"body": {
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
"__data__": {
"id": "chatcmpl-339",
"id": "chatcmpl-442",
"choices": [
{
"finish_reason": "length",
"index": 0,
"logprobs": null,
"message": {
"content": "I can guide you through the process, but please note that this is not an official OpenAI API call. OpenAI's API terms and conditions prohibit using their models for malicious purposes.\n\nTo test a model like \"text-temperature\" with a temperature of 0 (i.e., no noise or randomness), we'll need to use a third-party library that connects to the OpenAI API. One such library is `transformers`.\n\nFirst, you need to install the `transformers` and `",
"content": "I can guide you on how to use the `test-temperature` parameter with OpenAI's API, but please note that using a temperature of 0 may not produce meaningful results. Temperature is a hyperparameter that controls the level of randomness in the model's output.\n\nOpenAI's API uses a variant of the GPT-3 model, which is trained on a large corpus of text data. The `test-temperature` parameter allows you to adjust the level of randomness in the model's output",
"refusal": null,
"role": "assistant",
"annotations": null,
@ -39,7 +39,7 @@
}
}
],
"created": 1754510065,
"created": 1756921254,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion",
"service_tier": null,

6
tests/integration/recordings/responses/3c0bf9ba81b2.json
View file

@ -20,14 +20,14 @@
"body": {
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
"__data__": {
"id": "chatcmpl-651",
"id": "chatcmpl-334",
"choices": [
{
"finish_reason": "length",
"index": 0,
"logprobs": null,
"message": {
"content": "I'm ready to help",
"content": "It looks like we've",
"refusal": null,
"role": "assistant",
"annotations": null,
@ -37,7 +37,7 @@
}
}
],
"created": 1755294941,
"created": 1756921086,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion",
"service_tier": null,

30
tests/integration/recordings/responses/3c3f13cb7794.json
View file

@ -21,7 +21,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.338232Z",
"created_at": "2025-09-03T17:36:18.136699Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -39,7 +39,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.39419Z",
"created_at": "2025-09-03T17:36:18.177622Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -57,7 +57,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.445346Z",
"created_at": "2025-09-03T17:36:18.218104Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -75,7 +75,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.496701Z",
"created_at": "2025-09-03T17:36:18.258837Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -93,7 +93,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.546804Z",
"created_at": "2025-09-03T17:36:18.299715Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -111,7 +111,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.601009Z",
"created_at": "2025-09-03T17:36:18.341602Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -129,7 +129,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.652788Z",
"created_at": "2025-09-03T17:36:18.385504Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -147,7 +147,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.703325Z",
"created_at": "2025-09-03T17:36:18.429427Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -165,7 +165,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.754033Z",
"created_at": "2025-09-03T17:36:18.473547Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -183,7 +183,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.804654Z",
"created_at": "2025-09-03T17:36:18.516327Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -201,15 +201,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:11.854841Z",
"created_at": "2025-09-03T17:36:18.559332Z",
"done": true,
"done_reason": "stop",
"total_duration": 652371000,
"load_duration": 42086042,
"total_duration": 628034000,
"load_duration": 116384417,
"prompt_eval_count": 26,
"prompt_eval_duration": 78000000,
"prompt_eval_duration": 87798792,
"eval_count": 11,
"eval_duration": 531000000,
"eval_duration": 423189583,
"response": "",
"thinking": null,
"context": null

14
tests/integration/recordings/responses/3ca695048bee.json
View file

@ -1,7 +1,7 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/v1/v1/chat/completions",
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
"headers": {},
"body": {
"model": "llama3.2:3b-instruct-fp16",
@ -39,7 +39,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-490",
"id": "chatcmpl-3",
"choices": [
{
"delta": {
@ -50,7 +50,7 @@
"tool_calls": [
{
"index": 0,
"id": "call_rolv1ozt",
"id": "call_3kigugt3",
"function": {
"arguments": "{\"city\":\"Tokyo\"}",
"name": "get_weather"
@ -64,7 +64,7 @@
"logprobs": null
}
],
"created": 1754081852,
"created": 1756921361,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -75,7 +75,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-490",
"id": "chatcmpl-3",
"choices": [
{
"delta": {
@ -85,12 +85,12 @@
"role": "assistant",
"tool_calls": null
},
"finish_reason": "stop",
"finish_reason": "tool_calls",
"index": 0,
"logprobs": null
}
],
"created": 1754081852,
"created": 1756921361,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,

422
tests/integration/recordings/responses/3dff18060ebc.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"The secret string is foobazbar."
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.060630284,
0.06372823,
-0.059383437,
-0.010313639,
-0.11985778,
0.033409074,
0.056847293,
-0.0064553,
0.029896382,
-0.05037607,
0.015193001,
-0.0634204,
0.015119892,
-0.08354324,
0.0092577925,
0.044272587,
-0.024397198,
-0.05100177,
-0.028086444,
-0.07390362,
0.07088186,
0.08101153,
0.006050408,
-0.043090094,
0.010714593,
-0.01581376,
0.0351736,
0.06538307,
0.03639655,
-0.05625738,
0.073681176,
0.04730274,
0.067169026,
-0.01207242,
-0.018193275,
0.0042488067,
0.029168725,
0.0067459582,
0.037927665,
0.0024767139,
0.014044963,
0.022671249,
-0.090508185,
0.041952047,
-0.07933115,
0.031992197,
-0.038355146,
0.037013844,
-0.0036946274,
-0.016986867,
0.03696087,
-0.07697335,
-0.020080294,
0.07733012,
0.04521822,
-0.007816803,
-0.0058926586,
0.009962128,
0.033492323,
0.09000152,
0.016161384,
0.036999356,
-0.039193578,
-0.010969346,
0.023929566,
-0.03698458,
-0.008227196,
0.018780757,
-0.0006967325,
-0.062018193,
-0.030388007,
-0.037649162,
-0.04654288,
0.038450293,
-0.010377299,
-0.032971557,
0.013547814,
-0.059036925,
0.0630603,
0.0159564,
-0.04845087,
-0.069917254,
-0.022502322,
0.04408022,
0.03618941,
0.060470726,
-0.04313285,
0.028797466,
0.0062393937,
0.01027349,
-0.078714885,
-0.091531575,
0.04391341,
0.013202597,
-0.0037814155,
0.0102497,
0.020225797,
0.05634384,
-0.09700619,
0.06577961,
0.047118917,
0.01876648,
0.12445029,
-0.06447121,
-0.012632697,
0.016056264,
0.08604982,
0.024878234,
0.10627678,
-0.043176394,
-0.046339765,
-0.03149599,
-0.001784808,
-0.023469802,
-0.05079461,
0.0046657966,
0.043237828,
0.057146583,
-0.065833576,
0.032975562,
-0.028763266,
0.037831448,
0.00017829033,
0.043322463,
-0.13265091,
0.0263673,
-0.04247752,
-3.3340873e-33,
-0.0022191573,
0.050657377,
0.028066125,
-0.033898965,
-0.0045730886,
-0.034653578,
-0.08628417,
0.043108672,
0.01022734,
0.044009056,
-0.03020062,
-0.0936044,
-0.06522928,
-0.059762992,
0.037560984,
-0.025942331,
-0.06655938,
0.0043691625,
0.018846871,
-0.035582166,
0.02240012,
0.08943218,
0.033568345,
-0.11379316,
0.03822112,
-0.044403847,
0.10261262,
-0.07330182,
0.089390896,
0.056668896,
-0.009407597,
-0.0646505,
0.016652016,
0.007326742,
0.005187682,
0.0051324354,
-0.013595071,
-0.04918112,
-0.06672084,
0.010838405,
0.04638185,
-0.11490209,
-0.055054087,
0.040443793,
-0.032746885,
0.03498173,
-0.023567867,
-0.012213799,
0.048050664,
0.01159698,
0.007860181,
0.03801084,
-0.027765153,
0.003296162,
-0.0033349432,
0.006083357,
0.03200884,
0.048306234,
0.013800832,
0.036165927,
-0.022672432,
0.09197581,
0.029846204,
0.08112345,
-0.08677228,
-0.028041098,
0.0556574,
-0.030357547,
-0.016538681,
0.031826265,
-0.07586954,
-0.009915978,
0.028101236,
0.002207158,
-0.10496646,
-0.023673821,
-0.024204832,
-0.0003132271,
0.0016462951,
-0.037603874,
0.025533162,
-0.05221861,
0.021656586,
0.099111386,
-0.06896361,
-0.018568028,
0.07245527,
-0.10582686,
-0.08505038,
-0.029969748,
-0.015717981,
-0.056855034,
-0.02698479,
-0.06410572,
0.0057078917,
1.2902391e-33,
0.05490771,
-0.036417797,
-0.0023541928,
-0.03591478,
0.106852315,
-0.04931468,
0.037884213,
0.050633065,
-0.083874516,
-0.018756155,
0.0036251817,
0.028974183,
-0.0027879397,
-0.036439158,
0.11148004,
0.051007163,
0.040258586,
0.09245398,
-0.01367112,
-0.070999645,
-0.043213032,
-0.060117763,
-0.03019449,
0.009107182,
-0.044254936,
0.04843456,
0.117205575,
-0.009833911,
0.0023962231,
0.09339494,
-0.059902366,
0.0101377955,
-0.03777244,
-0.04344207,
-0.14677393,
-0.022666233,
-0.008934328,
-0.02157697,
-0.021902358,
-0.06611372,
0.016243221,
0.062620856,
0.01056146,
0.04721975,
-0.087221384,
0.009420561,
-0.017691165,
-0.03847053,
0.010398396,
0.022942957,
0.099518456,
-0.021421565,
0.0016765085,
-0.039359514,
0.01641369,
0.039669517,
-0.119695365,
0.009885617,
0.003855461,
0.018273395,
-0.0454586,
0.0020496584,
0.024263415,
0.016978405,
0.06884217,
-0.027432522,
-0.01813802,
0.053840507,
-0.028815664,
-0.045221787,
0.11472852,
0.019796453,
-0.05785514,
0.016556906,
-0.07362942,
0.04025756,
-0.01510899,
0.0067040483,
-0.049666926,
0.045941774,
0.077951804,
-0.042951427,
0.021852365,
0.063826546,
0.08110754,
-0.070652775,
-0.03245094,
0.09259784,
-0.020451743,
0.0701599,
-0.020740295,
0.09339449,
-0.051164806,
0.039440546,
0.02560772,
-1.6767814e-08,
0.001529873,
0.0080792755,
-0.017666567,
-0.034070052,
0.06805411,
0.07387949,
-0.07592055,
-0.11369049,
-0.022008128,
0.009088418,
0.03108134,
-0.0056734695,
-0.0462051,
0.0037219985,
0.013269294,
-0.03213892,
-0.05557376,
-0.010602884,
0.006751397,
-0.025462827,
-0.0836812,
0.08886153,
0.005159859,
-0.051621262,
-0.051873572,
0.039706588,
-0.042155124,
0.057125967,
0.088910565,
0.049736783,
0.04144574,
0.094677895,
-0.037107926,
-0.06845684,
-0.061673928,
0.09891817,
-0.05952751,
-0.0331722,
-0.026014913,
0.077612035,
0.056150436,
0.010709955,
0.018974187,
0.056079865,
-0.041700333,
-0.02731697,
0.10184176,
-0.036189064,
-0.029914921,
-0.043333948,
0.043660097,
0.018800316,
-0.0042763646,
0.055898346,
-0.0034344571,
0.060258396,
-0.1337251,
0.008184424,
-0.031549457,
0.022398692,
0.037932154,
0.024529235,
0.068037644,
0.07021777
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 9,
"total_tokens": 9
}
}
},
"is_streaming": false
}
}

422
tests/integration/recordings/responses/417020320684.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"Python programming language"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.063880146,
0.013411989,
-0.054502595,
0.01193493,
-0.074262686,
-0.13344447,
0.04294062,
0.045387108,
-0.06949706,
-0.035939943,
0.01200873,
0.0068830596,
0.08886977,
0.0026030506,
0.032482542,
-0.007821568,
-0.05044649,
0.006662123,
0.027794942,
-0.12791364,
0.00062353734,
0.045270294,
-0.03605076,
0.044243146,
0.0129354475,
-0.0092799105,
0.011904844,
0.026060482,
0.020055141,
-0.03368774,
-0.028043076,
0.087557025,
0.059002083,
0.053893365,
0.02027196,
0.06840361,
-0.03180594,
-0.087597735,
-0.11277839,
0.022651086,
-0.09037903,
-0.0033202847,
-0.040132593,
-0.034084503,
-0.032953303,
0.02925268,
-0.03903928,
0.04551951,
-0.0331016,
-0.006518362,
-0.09629851,
-0.011739161,
-0.052575007,
-0.064773224,
0.031043475,
-0.012586444,
0.09737276,
0.005224713,
-0.035071153,
-0.1404299,
-0.06678175,
0.03654573,
-0.039277818,
0.07014256,
-0.0010227569,
-0.026846789,
-0.0175696,
0.03044068,
0.06403526,
-0.031643596,
-0.14598879,
-0.045400888,
-0.018469285,
0.06689445,
0.030553635,
-0.12255281,
0.061046645,
-0.05678168,
-0.005118667,
-0.0087622,
0.006514719,
-0.016424034,
-0.033650044,
0.08491301,
-0.00029260007,
-0.07339515,
0.038627055,
0.15695965,
0.010035773,
0.025318887,
-0.0021428047,
-0.04613549,
0.06244243,
-0.019905778,
-0.05471386,
0.09796629,
0.0384793,
-0.072424814,
-0.038704097,
0.07158691,
0.007360897,
-0.05120446,
0.0313513,
-0.032230332,
0.039326303,
-0.009643992,
0.069905065,
-0.052026685,
0.049440835,
-0.04272916,
-0.0037707465,
-0.04155246,
-0.0561972,
-0.03340213,
0.05105359,
0.038616214,
-0.0029470131,
0.08188407,
-0.0035886324,
0.04530431,
0.0068888925,
0.016499842,
0.016347302,
0.007283021,
-0.021663606,
-0.0046215886,
-0.007931065,
-4.1536508e-33,
-0.045777988,
-0.050903402,
-0.038634304,
0.0100991195,
0.070007294,
-0.025182785,
0.1050647,
-0.0049731904,
-0.064141616,
-0.047639705,
0.012718577,
0.05198462,
-0.016051587,
0.08170543,
0.024008816,
-0.020879291,
0.045706064,
0.091577366,
0.02512945,
0.019055998,
0.048144504,
0.097951256,
0.034154113,
0.03543114,
0.011410896,
-0.043446988,
-0.0041784984,
-0.05564714,
0.01147717,
0.0071039577,
-0.06426582,
-0.020623188,
-0.0045247558,
-0.012943628,
0.02658834,
-0.012385487,
0.008399212,
-0.06824828,
0.04683057,
-0.04165085,
-0.025662417,
-0.0038799767,
0.05007075,
-0.008117481,
-0.023308154,
0.023914568,
0.0015741173,
0.046142872,
-0.06898886,
0.041611847,
0.0045286645,
-0.047628563,
0.054236773,
0.06972688,
-0.016889753,
0.04806098,
0.012714234,
0.0022186628,
-0.006355918,
-0.031550523,
0.023726372,
0.06859327,
0.077228814,
-0.01227583,
0.03901903,
0.034360897,
0.03032876,
0.058690928,
0.08030179,
0.06976231,
-0.09047136,
0.02376998,
-0.008751518,
0.038334776,
-0.02751323,
0.023137644,
0.027101006,
-0.08135271,
-0.010334998,
0.04730408,
-0.02033998,
-0.026008504,
-0.017415512,
-0.0035714875,
-0.018727385,
-0.037389226,
0.041064497,
0.05317889,
-0.0055602547,
-0.058561854,
-0.072036326,
-0.075019896,
0.04825644,
0.011348427,
-0.02259257,
1.3515749e-33,
0.006240622,
0.031606406,
-0.036119435,
-0.0016494404,
-0.08255665,
-0.06069396,
0.059934463,
0.014492232,
0.059514895,
0.027053975,
-0.011601325,
-0.057609312,
0.10365583,
-0.002784741,
0.07693759,
0.019432511,
-0.052210074,
0.015158053,
-0.0012768542,
0.027789148,
-0.115292676,
0.047323048,
-0.07599195,
-0.074344486,
-0.029194841,
-0.020079462,
-0.034749795,
-0.05769437,
-0.0301632,
0.04749987,
0.012206333,
0.011497502,
-0.051970575,
0.05972769,
0.03281016,
0.0013676677,
0.057720944,
-0.041179247,
-0.02150875,
-0.0067487382,
0.1419711,
0.05795878,
0.010094941,
0.09603845,
0.014521089,
0.02133803,
-0.07551916,
0.07887724,
-0.04273237,
-0.06601746,
-0.038729392,
-0.008161129,
0.015012324,
-0.049418066,
-0.037083283,
-0.02378242,
0.03743137,
0.008194503,
-0.086978436,
-0.05960285,
-0.07732487,
-0.056507926,
0.029065313,
0.0073954053,
-0.077878684,
0.0026059505,
-0.10405392,
-0.04738624,
-0.015872862,
-0.11591199,
0.09724705,
0.0049243565,
-0.010273523,
0.0066429917,
-0.060295314,
0.02550513,
-0.052950058,
-0.0038489713,
-0.050250847,
0.07679287,
0.046089787,
0.007386997,
0.0046740095,
0.07385862,
-0.07792065,
0.0013675193,
0.013730894,
0.05658653,
0.021934126,
0.007195913,
0.0076705213,
0.10221154,
0.060060997,
0.036779005,
-0.037765697,
-1.187368e-08,
-0.00885571,
0.01760442,
0.062224448,
0.032051455,
-0.011581793,
0.051908698,
-0.011685676,
-0.06391574,
-0.029866237,
0.03258576,
0.0055078953,
-0.012040446,
-0.054406017,
-0.056690563,
-0.030638037,
0.14276367,
0.028526368,
-0.028743364,
0.019917691,
0.025652615,
0.073813364,
-0.0066998666,
0.0061508445,
0.09610696,
-0.08799916,
-0.0089272335,
0.03823298,
0.04832936,
0.018829934,
-0.10534708,
0.048226915,
-0.02225069,
0.020491786,
0.014641141,
0.030794447,
-0.029119467,
0.008283775,
-0.04506887,
0.0025344177,
0.021756247,
-0.008108281,
0.00904927,
-0.013340866,
-0.014037631,
0.06845187,
0.045173325,
-0.034587316,
-0.07275669,
-0.004159724,
-0.058231864,
-0.033032075,
0.0040235794,
-0.019985583,
-0.020122562,
0.055365406,
0.10250875,
-0.10799118,
-0.013780294,
-0.009652406,
0.015592658,
-0.031221472,
0.1329332,
0.15243866,
-0.022426173
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 3,
"total_tokens": 3
}
}
},
"is_streaming": false
}
}

422
tests/integration/recordings/responses/4420515208a8.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"What is the secret string?"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.07473014,
0.08137506,
-0.06463602,
0.011821943,
-0.07454815,
0.021821007,
0.077573344,
0.012804661,
0.05853777,
-0.014141324,
0.053993534,
-0.026554074,
-0.018055506,
-0.060447972,
-0.019253474,
-0.006501444,
-0.047272332,
-0.048944764,
-0.090516366,
-0.06656194,
0.09287066,
0.02129739,
-0.013401809,
-0.006629013,
0.0079892,
0.016818035,
0.03971694,
0.021875564,
0.014873574,
-0.039426163,
0.025255844,
-0.036836684,
0.016627828,
0.008789532,
-0.053503897,
0.03616121,
-0.034633957,
-0.009877797,
0.064843215,
-0.01517806,
0.020897496,
-0.07135096,
-0.008519908,
0.05118655,
-0.062102985,
0.059486073,
-0.047937352,
0.07045817,
-0.024867272,
-0.010756205,
0.06538509,
-0.03693754,
-0.08240387,
0.08169191,
0.017090658,
0.012944557,
-0.047139525,
0.0025796075,
0.008701712,
0.099866174,
0.04969699,
-0.025922626,
-0.017354922,
0.03395182,
0.038391408,
-0.054247838,
0.008610521,
-0.04077977,
0.0265637,
-0.07186012,
-0.019953186,
-0.041191205,
-0.07246228,
0.00041248833,
0.018758524,
0.023036895,
0.01662864,
-0.06335885,
0.03495032,
0.050063577,
0.00043262896,
-0.06176693,
0.0062733325,
0.11142063,
0.0040838965,
0.085737824,
0.023284689,
0.05699812,
-0.03149832,
-0.013344509,
-0.045138564,
-0.117300816,
0.016063986,
-0.016894838,
-0.028934335,
0.03575864,
-0.05156192,
0.032958068,
-0.11266628,
0.06640015,
0.037839692,
0.022948038,
0.058071073,
-0.039643735,
-0.03247236,
0.017690921,
-0.005001274,
0.019046135,
0.07745316,
-0.020402163,
-0.020310633,
-0.009519755,
0.0031459313,
-0.0045639877,
-0.029116316,
0.033835515,
0.00050839526,
0.06419946,
0.010721198,
0.124151744,
-0.0053820186,
0.00491648,
-0.059696514,
0.029483523,
-0.13409872,
0.016187217,
-0.048092023,
-6.6084764e-33,
0.012305612,
0.060384244,
0.036461998,
-0.035974216,
-0.04197416,
0.012333701,
-0.084805995,
0.012502633,
0.02794982,
0.0861082,
-0.030791838,
-0.061355945,
-0.0009604986,
-0.0252044,
0.045444816,
-0.027590565,
-0.009594973,
0.006712001,
0.043692384,
-0.021483036,
0.003300438,
0.11860881,
0.047044385,
-0.1348901,
0.025469579,
-0.01029819,
0.0022393467,
-0.061863262,
0.10386513,
0.018658707,
-0.0017492755,
-0.051914047,
0.046442248,
0.03761067,
0.033752125,
0.006650237,
0.022015076,
-0.07834835,
-0.008209136,
0.027432231,
0.017393896,
-0.07524756,
0.006497012,
0.027272953,
0.0005804994,
-0.010941825,
-0.020050043,
-0.00012092298,
0.013705002,
0.004699541,
0.022770848,
0.015477994,
-0.0142482165,
-0.013953546,
0.015865315,
-0.023075614,
0.03379947,
-0.039221376,
-0.043229815,
0.02998769,
-0.01652291,
0.06981088,
0.04606923,
0.05332633,
-0.055300076,
0.02511626,
0.014049543,
-0.09398743,
0.03590562,
0.029452223,
-0.13200304,
-0.005059034,
-0.03784268,
-0.03180819,
-0.095502876,
-0.027853556,
0.0024331037,
-0.007881495,
0.058296,
-0.031999517,
-0.06077097,
-0.023381822,
-0.00048603877,
0.13765746,
-0.060579,
-0.008109843,
-0.034873307,
-0.1024547,
-0.009072849,
-0.018931676,
-0.0016711762,
-0.07710289,
-0.043332253,
-0.03619527,
0.03958017,
3.0217083e-33,
0.0050329794,
0.00016030145,
-0.063078895,
0.012225751,
0.10637338,
0.015972024,
0.006653195,
0.01880781,
-0.04708357,
0.045863643,
0.0076015075,
0.03243478,
0.032097474,
-0.020893326,
0.10697852,
0.0075498912,
0.036074348,
0.1462344,
0.03779065,
-0.043190572,
-0.02176097,
-0.009340132,
-0.06983617,
0.015578788,
0.021121953,
0.030661412,
0.08434581,
-0.09288574,
0.008169474,
0.078080945,
-0.081626564,
0.011895231,
0.017099649,
0.0040119104,
-0.14145434,
0.0040375097,
0.046316408,
0.008959473,
-0.0056506568,
-0.055587813,
0.028007837,
0.055937108,
0.062269785,
0.08602392,
-0.12157818,
0.021943888,
-0.0050934856,
0.029819332,
-0.012127162,
0.048801802,
0.06409215,
-0.041438665,
0.01809265,
-0.028214281,
-0.0213588,
0.05564267,
-0.1547868,
0.027465124,
0.018855799,
0.04327939,
0.011500479,
0.017364705,
-0.023216385,
0.051007293,
0.02946264,
0.012533944,
-0.04542834,
-0.002238765,
-0.05611544,
-0.0789272,
0.07960444,
-0.020431034,
-0.0762138,
0.011588508,
-0.035614885,
-0.04803985,
-0.06607436,
-0.057365946,
-0.040188126,
0.07176218,
0.03135825,
0.02303279,
-0.023997622,
0.023614945,
0.09607302,
-0.06843066,
0.014260722,
0.08802569,
-0.037736766,
0.029445928,
-0.028643936,
0.10217973,
-0.0660917,
0.022864237,
0.042151757,
-1.4814046e-08,
0.030838449,
0.043877687,
-0.0245681,
-0.09818859,
0.056659035,
0.0929652,
-0.010337853,
-0.0983916,
0.018008571,
-0.0131424805,
0.026400762,
0.008793538,
-0.05285605,
-0.042175982,
0.030133193,
0.01710666,
-0.06242493,
-0.018753909,
-0.015986755,
-0.018400662,
-0.026477808,
0.010281372,
-0.030476814,
-0.084556945,
-0.05402664,
0.010030052,
0.029531356,
0.13555466,
0.033426728,
0.12098221,
0.040777553,
0.008206964,
-0.018235989,
-0.0568263,
-0.1289943,
0.12416113,
-0.053454727,
-0.038151894,
0.030221034,
0.019807614,
0.047819767,
0.029434063,
0.0015704447,
0.0611775,
-0.05557245,
-0.030236417,
0.10799873,
-0.07073352,
-0.08215229,
0.004518122,
-0.015573616,
-0.013696145,
-0.0023438279,
0.026377691,
-0.015769389,
0.016251203,
-0.04062322,
-0.013962793,
-0.08309221,
0.031991288,
0.049991824,
-0.0038595141,
0.07031122,
0.0049263495
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 6,
"total_tokens": 6
}
}
},
"is_streaming": false
}
}

4
tests/integration/recordings/responses/44a1d9de0602.json
View file

@ -20,7 +20,7 @@
"body": {
"__type__": "openai.types.chat.chat_completion.ChatCompletion",
"__data__": {
"id": "chatcmpl-987",
"id": "chatcmpl-507",
"choices": [
{
"finish_reason": "length",
@ -37,7 +37,7 @@
}
}
],
"created": 1755294921,
"created": 1756921150,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion",
"service_tier": null,

10
tests/integration/recordings/responses/44fb9cf5875f.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-07-31T17:59:42.166585642Z",
"created_at": "2025-09-03T17:41:49.581065Z",
"done": true,
"done_reason": "stop",
"total_duration": 9490295253,
"load_duration": 42349084,
"total_duration": 2391571708,
"load_duration": 182022958,
"prompt_eval_count": 20,
"prompt_eval_duration": 545470166,
"prompt_eval_duration": 74456583,
"eval_count": 51,
"eval_duration": 8901928284,
"eval_duration": 2134471458,
"response": "It seems like you're trying to test the system, but I'm not sure what specific functionality or feature you'd like to test. Could you please provide more context or clarify what you're looking for? I'll do my best to assist you!",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/48d2fb183a2a.json
View file

@ -67,15 +67,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:40.583477Z",
"created_at": "2025-09-03T17:36:40.283084Z",
"done": true,
"done_reason": "stop",
"total_duration": 3928481500,
"load_duration": 151903250,
"total_duration": 2900042958,
"load_duration": 83372125,
"prompt_eval_count": 259,
"prompt_eval_duration": 468000000,
"prompt_eval_duration": 352890750,
"eval_count": 60,
"eval_duration": 3306000000,
"eval_duration": 2462885208,
"response": "{\n \"first_name\": \"Michael\",\n \"last_name\": \"Jordan\",\n \"year_of_birth\": 1963,\n \"nba_stats\": {\n \"year_for_draft\": 1984,\n \"num_seasons_in_nba\": 15\n }\n}",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/50340cd4d253.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:14:19.298378Z",
"created_at": "2025-09-03T17:38:01.239743Z",
"done": true,
"done_reason": "stop",
"total_duration": 266786083,
"load_duration": 53820458,
"total_duration": 207264667,
"load_duration": 73437959,
"prompt_eval_count": 216,
"prompt_eval_duration": 192000000,
"prompt_eval_duration": 121657333,
"eval_count": 2,
"eval_duration": 17000000,
"eval_duration": 11348417,
"response": "safe",
"thinking": null,
"context": null

422
tests/integration/recordings/responses/5370751803dc.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"Python is a high-level programming language with code readability and fewer lines than C++ or Java"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.07642644,
0.0213101,
-0.03612849,
-0.0012144424,
-0.048599217,
-0.13194773,
-0.084226094,
0.059389386,
-0.0617182,
-0.009323243,
-0.08099486,
0.055514984,
0.052610602,
0.026061919,
0.063071534,
-0.062316332,
-0.065115415,
-0.022351492,
0.017378356,
-0.11605584,
-0.036349725,
0.0404155,
-0.0325302,
-0.01770141,
0.05722761,
0.012393438,
-0.018529164,
-0.030017126,
0.002365914,
0.0066701965,
-0.08862459,
0.0779319,
0.03702611,
0.029523117,
-0.01977821,
0.05424799,
-0.00074063655,
-0.08949148,
-0.05312112,
-0.012703181,
-0.08622611,
0.07689996,
-0.038602136,
-0.011616902,
-0.03234132,
-0.0073969415,
-0.024779495,
-0.067999884,
-0.03039565,
-0.025974417,
-0.09690519,
0.009931951,
-0.05362519,
-0.09107193,
-0.009222061,
-0.008804084,
0.048185978,
-0.003329437,
-0.0058579347,
-0.13306528,
-0.09721703,
0.013474277,
0.047286008,
0.06279936,
-0.01582815,
-0.03771013,
-0.01651892,
0.029905442,
0.09326656,
-0.06746783,
-0.13385954,
-0.020873511,
-0.02586237,
0.11623731,
0.030632136,
-0.10494776,
0.03905967,
-0.010701787,
-0.0014734551,
0.020711906,
0.0017687598,
0.027797814,
-0.078500465,
0.10791581,
0.02910256,
-0.05398749,
0.030513834,
0.07001416,
-0.034323946,
0.00986597,
0.034644563,
-0.04232179,
0.065106474,
0.026648693,
-0.032122962,
0.07616709,
0.020026332,
-0.030642457,
-0.07188906,
0.027189687,
-0.018678213,
-0.05416582,
0.07488992,
0.017753933,
0.03386007,
0.02414506,
0.09077034,
-0.052096054,
0.040722203,
-0.018450806,
-0.012474094,
-0.06403705,
-0.023205942,
-0.061878704,
0.053436812,
0.047876816,
-0.010608645,
0.07852118,
0.03579911,
0.027097313,
0.022424318,
-0.004912598,
-0.02455264,
0.003700777,
0.00039888592,
-0.008842094,
0.009365857,
2.05052e-34,
-0.03236592,
-0.024301885,
0.027186498,
0.021633558,
0.06519107,
-0.019539308,
0.05306087,
0.007985293,
-0.03927361,
-0.020062907,
0.008070545,
0.02382429,
0.015006528,
0.1128094,
0.06113956,
-0.011911169,
0.016901307,
0.045509744,
0.0013988831,
0.00907712,
0.01314859,
-0.012022324,
0.027043821,
0.0071581583,
0.022573117,
-0.013721936,
-0.004378743,
-0.0007087661,
0.033585846,
0.011227843,
-0.05136015,
-0.0739591,
-0.03094639,
0.01957863,
-0.010360539,
-0.0029881562,
-0.00480912,
-0.10446798,
0.034694213,
-0.02424012,
-0.047155295,
0.035451673,
0.037169226,
-0.016986743,
0.0056092087,
0.05057555,
-0.008601115,
0.0060349177,
-0.12273999,
0.036871877,
-0.022267655,
-0.009739047,
0.075974636,
0.08902226,
0.01647873,
0.044345584,
0.06792565,
0.06456903,
-0.050189856,
-0.0016995457,
-0.00090498856,
0.09925942,
0.09253569,
-0.011321612,
0.050309792,
0.07697773,
0.0100068,
0.101032645,
0.03268899,
0.06433435,
-0.044524822,
0.03860177,
-0.019314477,
0.037440598,
-0.0017394378,
0.011816814,
0.011359969,
-0.1040215,
0.06984421,
0.01910163,
-0.028409261,
-0.013704911,
0.048502754,
-0.015429918,
-0.03423058,
-0.055616368,
0.005001686,
0.026054256,
-0.0007700968,
-0.0041726283,
-0.0640977,
-0.05985385,
0.0813829,
0.014288322,
-0.038147252,
-2.1576616e-33,
-0.027279941,
-0.034765568,
-0.02465107,
0.026859807,
-0.090699576,
-0.045698144,
0.013666582,
0.002109106,
0.054007426,
0.032838397,
-0.029939773,
-0.058843046,
0.09825693,
0.03251322,
0.109977886,
0.020682266,
-0.0958973,
0.0005566991,
0.0018037638,
0.017544486,
-0.06843023,
0.06435102,
-0.050149646,
-0.048880838,
-0.027535524,
-0.014993001,
-0.1210176,
-0.04412877,
-0.011025324,
0.058610573,
-0.007498303,
0.038722932,
-0.07025986,
0.030281536,
0.055707317,
-0.001162887,
0.01707519,
-0.042081844,
-0.016578361,
-0.025714336,
0.117893435,
0.04196084,
0.064787276,
0.046081997,
0.014950138,
0.030026693,
-0.039077066,
0.087156676,
-0.012328571,
-0.035646956,
-0.048145168,
0.041394625,
0.038984135,
-0.025188481,
-0.028836856,
-0.02917782,
0.029690607,
0.051454436,
-0.08629761,
-0.06921346,
-0.07273269,
-0.05952071,
0.0050034616,
0.025693603,
-0.022103382,
0.024972659,
-0.09724792,
0.0062089814,
-0.04963219,
-0.13054384,
0.124669954,
-0.01361085,
-0.022798477,
0.039057832,
-0.07550591,
0.049364913,
0.0007779102,
0.004692535,
-0.040757872,
0.06355995,
0.110190175,
0.02015945,
-0.048807338,
0.05842704,
-0.066375315,
0.026938869,
-0.062775925,
-0.014049011,
0.023343485,
0.02358394,
-0.002172394,
0.07766165,
0.031056313,
0.020171564,
-0.020073414,
-2.4317085e-08,
0.020261949,
-0.008623839,
0.0621209,
-0.008334477,
0.02526615,
0.08902315,
-0.007958188,
-0.018911751,
-0.035572145,
0.06189234,
-0.017249323,
-0.030186126,
-0.10225455,
-0.06522741,
-0.004033112,
0.10897627,
-0.02168822,
-0.053784374,
0.011841631,
0.052263785,
0.058334205,
0.0052479547,
-0.06017166,
0.08723854,
-0.08275336,
-0.040676847,
0.065786876,
0.028317772,
-0.012168614,
-0.07196286,
0.014588226,
-0.03231537,
0.0028357722,
0.03868031,
0.055439528,
-0.015238348,
0.05482384,
-0.025080629,
-0.033771332,
0.0030752022,
-0.037511814,
0.015122315,
0.02292684,
0.012024873,
0.03559873,
0.006865039,
-0.04049267,
-0.049685854,
-0.05455341,
-0.073071465,
-0.024902396,
-0.002133957,
-0.013212662,
-0.06657236,
0.023245512,
0.046919,
-0.13278763,
-0.011092663,
-0.023939205,
0.043182902,
0.024406029,
0.06922961,
0.15658055,
0.017658537
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 21,
"total_tokens": 21
}
}
},
"is_streaming": false
}
}

34
tests/integration/recordings/responses/545d86510a80.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.59711Z",
"created_at": "2025-09-03T17:42:32.625862Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.671294Z",
"created_at": "2025-09-03T17:42:32.668885Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.736161Z",
"created_at": "2025-09-03T17:42:32.710947Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.809857Z",
"created_at": "2025-09-03T17:42:32.752286Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.883599Z",
"created_at": "2025-09-03T17:42:32.793309Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.942471Z",
"created_at": "2025-09-03T17:42:32.834578Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:38.999844Z",
"created_at": "2025-09-03T17:42:32.876536Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.050862Z",
"created_at": "2025-09-03T17:42:32.918807Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.104589Z",
"created_at": "2025-09-03T17:42:32.960101Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.158301Z",
"created_at": "2025-09-03T17:42:33.00196Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.210985Z",
"created_at": "2025-09-03T17:42:33.043876Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.263525Z",
"created_at": "2025-09-03T17:42:33.08756Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,15 +238,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:39.314455Z",
"created_at": "2025-09-03T17:42:33.12966Z",
"done": true,
"done_reason": "stop",
"total_duration": 914060542,
"load_duration": 63705209,
"total_duration": 648814958,
"load_duration": 75300875,
"prompt_eval_count": 408,
"prompt_eval_duration": 95000000,
"prompt_eval_duration": 66740291,
"eval_count": 13,
"eval_duration": 753000000,
"eval_duration": 505313125,
"response": "",
"thinking": null,
"context": null

46
tests/integration/recordings/responses/554de3cd986f.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.40585Z",
"created_at": "2025-09-03T17:37:51.805591Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.455647Z",
"created_at": "2025-09-03T17:37:51.850067Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.509581Z",
"created_at": "2025-09-03T17:37:51.892443Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.56592Z",
"created_at": "2025-09-03T17:37:51.934364Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.616979Z",
"created_at": "2025-09-03T17:37:51.978382Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.671413Z",
"created_at": "2025-09-03T17:37:52.019332Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.725494Z",
"created_at": "2025-09-03T17:37:52.060708Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.779905Z",
"created_at": "2025-09-03T17:37:52.102717Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.829791Z",
"created_at": "2025-09-03T17:37:52.143996Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.880729Z",
"created_at": "2025-09-03T17:37:52.185479Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.93338Z",
"created_at": "2025-09-03T17:37:52.227562Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:04.981714Z",
"created_at": "2025-09-03T17:37:52.270178Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,7 +238,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.036068Z",
"created_at": "2025-09-03T17:37:52.31151Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -256,7 +256,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.088069Z",
"created_at": "2025-09-03T17:37:52.35278Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -274,7 +274,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.144485Z",
"created_at": "2025-09-03T17:37:52.393954Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -292,7 +292,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.203042Z",
"created_at": "2025-09-03T17:37:52.435238Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -310,7 +310,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.257133Z",
"created_at": "2025-09-03T17:37:52.476197Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -328,7 +328,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.311623Z",
"created_at": "2025-09-03T17:37:52.517914Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -346,15 +346,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:05.370124Z",
"created_at": "2025-09-03T17:37:52.55904Z",
"done": true,
"done_reason": "stop",
"total_duration": 1532801458,
"load_duration": 213911041,
"total_duration": 971882292,
"load_duration": 116634209,
"prompt_eval_count": 376,
"prompt_eval_duration": 350000000,
"prompt_eval_duration": 99382958,
"eval_count": 19,
"eval_duration": 967000000,
"eval_duration": 755260750,
"response": "",
"thinking": null,
"context": null

30
tests/integration/recordings/responses/561746e1c8de.json
View file

@ -21,7 +21,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:49.18651486Z",
"created_at": "2025-09-03T17:36:20.465701Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -39,7 +39,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:49.370611348Z",
"created_at": "2025-09-03T17:36:20.507671Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -57,7 +57,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:49.557000029Z",
"created_at": "2025-09-03T17:36:20.549443Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -75,7 +75,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:49.746777116Z",
"created_at": "2025-09-03T17:36:20.590803Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -93,7 +93,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:49.942233333Z",
"created_at": "2025-09-03T17:36:20.631683Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -111,7 +111,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:50.126788846Z",
"created_at": "2025-09-03T17:36:20.672443Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -129,7 +129,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:50.311346131Z",
"created_at": "2025-09-03T17:36:20.713329Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -147,7 +147,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:50.501507173Z",
"created_at": "2025-09-03T17:36:20.754254Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -165,7 +165,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:50.692296777Z",
"created_at": "2025-09-03T17:36:20.795119Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -183,7 +183,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:50.878846539Z",
"created_at": "2025-09-03T17:36:20.836145Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -201,15 +201,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-15T20:24:51.063200561Z",
"created_at": "2025-09-03T17:36:20.877784Z",
"done": true,
"done_reason": "stop",
"total_duration": 33982453650,
"load_duration": 2909001805,
"total_duration": 612057417,
"load_duration": 97443583,
"prompt_eval_count": 341,
"prompt_eval_duration": 29194357307,
"prompt_eval_duration": 100914750,
"eval_count": 11,
"eval_duration": 1878247732,
"eval_duration": 413024250,
"response": "",
"thinking": null,
"context": null

10
tests/integration/recordings/responses/563b994bb7d1.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.25248Z",
"created_at": "2025-09-03T17:36:19.594923Z",
"done": true,
"done_reason": "stop",
"total_duration": 1344654917,
"load_duration": 200585375,
"total_duration": 988472417,
"load_duration": 117976625,
"prompt_eval_count": 326,
"prompt_eval_duration": 564000000,
"prompt_eval_duration": 451625542,
"eval_count": 11,
"eval_duration": 578000000,
"eval_duration": 418313417,
"response": "[get_weather(location=\"San Francisco, CA\")]",
"thinking": null,
"context": null

30
tests/integration/recordings/responses/5f5d16afadb4.json
View file

@ -21,7 +21,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.354888Z",
"created_at": "2025-09-03T17:36:19.808372Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -39,7 +39,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.427569Z",
"created_at": "2025-09-03T17:36:19.84991Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -57,7 +57,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.486244Z",
"created_at": "2025-09-03T17:36:19.892111Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -75,7 +75,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.540455Z",
"created_at": "2025-09-03T17:36:19.933857Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -93,7 +93,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.594439Z",
"created_at": "2025-09-03T17:36:19.975148Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -111,7 +111,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.649837Z",
"created_at": "2025-09-03T17:36:20.016641Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -129,7 +129,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.703358Z",
"created_at": "2025-09-03T17:36:20.058229Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -147,7 +147,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.7553Z",
"created_at": "2025-09-03T17:36:20.100222Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -165,7 +165,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.807251Z",
"created_at": "2025-09-03T17:36:20.143456Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -183,7 +183,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.857952Z",
"created_at": "2025-09-03T17:36:20.184657Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -201,15 +201,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:13.918522Z",
"created_at": "2025-09-03T17:36:20.226017Z",
"done": true,
"done_reason": "stop",
"total_duration": 647785042,
"load_duration": 26355584,
"total_duration": 598395375,
"load_duration": 129432167,
"prompt_eval_count": 326,
"prompt_eval_duration": 55000000,
"prompt_eval_duration": 50057334,
"eval_count": 11,
"eval_duration": 557000000,
"eval_duration": 418284791,
"response": "",
"thinking": null,
"context": null

422
tests/integration/recordings/responses/62aa454ea5f9.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"What inspires neural networks?"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.08566708,
-0.09559047,
0.044014607,
-0.015974598,
0.029406257,
0.07229597,
-0.010901963,
-0.023829829,
0.07381301,
-0.05698464,
-0.033780586,
0.051200844,
0.0050912783,
0.014317088,
-0.07878143,
-0.012908666,
-0.041628323,
0.06881713,
-0.10783476,
-0.04042705,
0.026262026,
-0.0019893218,
-0.011008084,
-0.0019646112,
0.004033132,
0.08881656,
0.014049165,
-0.018416086,
0.032621212,
-0.034692146,
0.07614942,
-0.014122101,
-0.024901746,
0.03755059,
-0.10197354,
0.054705318,
-0.022539826,
0.024209768,
0.011698194,
-0.008956377,
-0.050146304,
0.0026327297,
0.055942897,
0.009974366,
0.12796965,
-0.025006283,
0.024338534,
-0.024487961,
-0.0022703854,
-0.024687177,
-0.10482094,
-0.05994297,
-0.055200897,
0.0152664175,
0.03496896,
0.052624088,
-0.0006445885,
0.06637695,
-0.031790398,
-0.007308742,
-0.0050764186,
-0.042508755,
-0.04089097,
0.020062948,
0.038683955,
0.022463562,
-0.02866933,
0.053370677,
0.022435635,
0.01934692,
0.12264713,
0.023911418,
-0.037264284,
0.0059156846,
0.05235448,
0.054004095,
0.08022169,
-0.010992806,
0.029295033,
-0.0672064,
-0.00021147476,
-0.050584126,
-0.0095251575,
0.04616498,
0.078677796,
0.01416309,
-0.033226117,
0.0018380182,
-0.06667651,
-0.020977372,
-0.017116925,
-0.04396714,
-0.05969979,
-0.07344942,
-0.03985366,
-0.030863814,
-0.019918729,
-0.1075161,
-0.026654154,
0.0689854,
-0.0049292273,
0.026645623,
0.018879393,
0.022113768,
0.064208575,
-0.053153764,
0.06160797,
0.014026719,
0.11772326,
-0.051769163,
-0.07634968,
0.03090975,
-0.038558383,
-0.025260162,
0.039262023,
-0.061449137,
0.008389126,
0.016175874,
0.032293033,
0.06679397,
-0.06503257,
0.014676881,
-0.038542666,
0.018718671,
-0.030111106,
-0.028481327,
-0.14707623,
-3.455443e-33,
-0.048577547,
-0.024983348,
0.071679614,
0.035652317,
0.07931413,
-0.07811974,
0.023085583,
-0.047467884,
0.08872273,
-0.0010074769,
-0.11320135,
0.091322996,
0.023978539,
0.11368158,
0.042203873,
-0.05773289,
-0.074543044,
-0.0021036167,
-0.051522236,
-0.050925426,
-0.0016557347,
0.030671587,
0.045119714,
-0.03974729,
-0.05871358,
-0.030611658,
0.0017253247,
0.009114429,
-0.013763352,
0.023424039,
0.0017495834,
0.046633217,
-0.07230643,
-0.027882291,
0.016182518,
0.044456217,
-0.004326421,
-0.061798126,
0.0697968,
0.031249145,
-0.013697079,
-0.007417679,
0.031665757,
-0.02367961,
0.07153089,
0.023938214,
0.009729952,
0.0071919435,
-0.03235391,
-0.04955071,
-0.050248373,
0.02151118,
0.015327139,
-0.0674203,
0.06544387,
-0.025547959,
0.03207046,
0.02038825,
0.0112230005,
0.00019493286,
-0.023462659,
-0.004949742,
-0.014066955,
0.0014178518,
0.059315395,
0.039931085,
-0.032498423,
-0.023698896,
0.05445033,
0.064231694,
-0.034013335,
0.08745776,
-0.080473825,
-0.090545714,
-0.065398656,
-8.2386265e-05,
-0.021441188,
-0.0684535,
-0.029121745,
0.034134887,
-0.07799698,
-0.05388711,
-0.035591345,
0.044826802,
-0.040090464,
0.07972004,
0.026058797,
-0.08184859,
0.0018106091,
-0.027676936,
-0.04312832,
-0.042090744,
0.08336437,
-0.049453646,
-0.0902778,
2.6716498e-33,
-0.091911495,
0.02641473,
-0.07022486,
0.075562105,
0.03900905,
0.027913846,
-0.05444872,
-0.036666486,
-0.048225258,
0.07551892,
0.046452336,
0.025874302,
0.052248206,
-0.00018527219,
0.010575236,
-0.040591337,
-0.028484622,
-0.020559357,
0.08882296,
-0.06755767,
0.04941752,
0.13231009,
-0.06998129,
-0.040112328,
0.044030365,
0.034218542,
-0.08650528,
0.05746921,
-0.0075130556,
0.049070083,
-0.0148686,
-0.018103259,
-0.020280316,
0.038828347,
0.022253176,
0.13486238,
0.06899369,
-0.002589861,
-0.016430879,
0.0033818923,
0.017275693,
0.013614936,
0.044220798,
0.049155377,
-0.008259856,
-0.046575654,
-0.043921605,
0.04156687,
-0.035468902,
0.042837795,
0.03131579,
0.017961076,
-0.026213305,
-0.05458616,
-0.04259084,
-0.004110002,
0.029035388,
0.0010451805,
0.09044077,
0.014110149,
-0.068820216,
-0.07098938,
0.020328037,
0.00433692,
-0.046977337,
0.016492791,
-0.028396707,
0.104340956,
0.002814702,
-0.08339559,
0.037326302,
0.058929898,
0.0376423,
0.09580634,
-0.12376848,
-0.054060236,
-0.014485116,
0.0013106487,
-0.04537336,
-0.0899294,
0.001730278,
-0.05520831,
0.000568523,
0.00053380145,
0.07856981,
0.104590714,
0.00355283,
0.008365939,
0.04291482,
0.010064388,
0.025177509,
0.05732803,
-0.023061136,
0.054399785,
-0.049828697,
-1.3290186e-08,
-0.0539168,
0.08074109,
0.03397028,
0.024365881,
0.0906225,
-0.07162824,
0.07550329,
0.017278913,
-0.061226364,
-0.03298407,
0.07829606,
0.03967995,
-0.036696997,
0.02665964,
0.1000655,
-0.014426734,
0.020708792,
-0.039230846,
0.0085029,
-0.0012509917,
0.06740856,
0.013992665,
-0.054007422,
-0.016785627,
0.07651403,
-0.035508703,
-0.050085396,
0.08382383,
-0.009957674,
0.08140875,
0.019287178,
0.049911316,
0.0022236605,
-0.07807412,
0.019454133,
0.111560374,
-0.01269702,
-0.06466137,
-0.09346588,
-0.050038446,
-0.042178612,
0.0599713,
0.034831088,
-0.014957726,
0.014484159,
-0.022619838,
0.06916277,
-0.088544875,
0.021478733,
0.01378541,
-0.0075770007,
0.027888266,
0.015526889,
0.0052174823,
0.010616002,
-0.022908956,
-0.02535865,
-0.04139556,
-0.08375561,
0.092626974,
0.051755503,
0.09296614,
0.011223383,
-0.016759252
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 6,
"total_tokens": 6
}
}
},
"is_streaming": false
}
}

10
tests/integration/recordings/responses/6906a6e71988.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama-guard3:1b",
"created_at": "2025-08-01T23:14:18.886381Z",
"created_at": "2025-09-03T17:38:00.98692Z",
"done": true,
"done_reason": "stop",
"total_duration": 488566500,
"load_duration": 113477291,
"total_duration": 332473583,
"load_duration": 90611333,
"prompt_eval_count": 317,
"prompt_eval_duration": 361000000,
"prompt_eval_duration": 229691000,
"eval_count": 2,
"eval_duration": 12000000,
"eval_duration": 11571291,
"response": "safe",
"thinking": null,
"context": null

48
tests/integration/recordings/responses/6cc063bbd7d3.json
View file

@ -21,7 +21,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:55.9885Z",
"created_at": "2025-09-03T17:42:17.402486Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -39,7 +39,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.054143Z",
"created_at": "2025-09-03T17:42:17.444334Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -57,7 +57,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.117658Z",
"created_at": "2025-09-03T17:42:17.484625Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -75,7 +75,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.179422Z",
"created_at": "2025-09-03T17:42:17.525063Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -93,7 +93,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.240328Z",
"created_at": "2025-09-03T17:42:17.565015Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -111,7 +111,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.295992Z",
"created_at": "2025-09-03T17:42:17.60499Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -129,7 +129,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.355683Z",
"created_at": "2025-09-03T17:42:17.64509Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -147,7 +147,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.412176Z",
"created_at": "2025-09-03T17:42:17.685566Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -165,7 +165,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.466952Z",
"created_at": "2025-09-03T17:42:17.725855Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -183,7 +183,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.517222Z",
"created_at": "2025-09-03T17:42:17.766056Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -201,7 +201,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.570491Z",
"created_at": "2025-09-03T17:42:17.806415Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -219,7 +219,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.623189Z",
"created_at": "2025-09-03T17:42:17.847273Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -237,7 +237,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.679221Z",
"created_at": "2025-09-03T17:42:17.888576Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -255,7 +255,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.731373Z",
"created_at": "2025-09-03T17:42:17.928952Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -273,7 +273,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.781364Z",
"created_at": "2025-09-03T17:42:17.969744Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -291,7 +291,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.831951Z",
"created_at": "2025-09-03T17:42:18.010869Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -309,7 +309,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.888381Z",
"created_at": "2025-09-03T17:42:18.051109Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -327,7 +327,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.943539Z",
"created_at": "2025-09-03T17:42:18.093266Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -345,7 +345,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:56.997422Z",
"created_at": "2025-09-03T17:42:18.135749Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -363,15 +363,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:57.056259Z",
"created_at": "2025-09-03T17:42:18.176649Z",
"done": true,
"done_reason": "stop",
"total_duration": 1289815458,
"load_duration": 119745583,
"total_duration": 907420000,
"load_duration": 66756750,
"prompt_eval_count": 26,
"prompt_eval_duration": 98000000,
"prompt_eval_duration": 62900875,
"eval_count": 20,
"eval_duration": 1071000000,
"eval_duration": 777306958,
"response": "",
"thinking": null,
"context": null

34
tests/integration/recordings/responses/6d35c91287e2.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.362667Z",
"created_at": "2025-09-03T17:38:03.549266Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.427435Z",
"created_at": "2025-09-03T17:38:03.592203Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.484198Z",
"created_at": "2025-09-03T17:38:03.63417Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.537031Z",
"created_at": "2025-09-03T17:38:03.677268Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.591198Z",
"created_at": "2025-09-03T17:38:03.719768Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.643336Z",
"created_at": "2025-09-03T17:38:03.762204Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.698589Z",
"created_at": "2025-09-03T17:38:03.80404Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.752904Z",
"created_at": "2025-09-03T17:38:03.845678Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.804Z",
"created_at": "2025-09-03T17:38:03.887086Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.855633Z",
"created_at": "2025-09-03T17:38:03.928422Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.906918Z",
"created_at": "2025-09-03T17:38:03.969641Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:22.958729Z",
"created_at": "2025-09-03T17:38:04.011212Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,15 +238,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:23.011279Z",
"created_at": "2025-09-03T17:38:04.052626Z",
"done": true,
"done_reason": "stop",
"total_duration": 793500292,
"load_duration": 55339750,
"total_duration": 731936583,
"load_duration": 147334791,
"prompt_eval_count": 417,
"prompt_eval_duration": 83000000,
"prompt_eval_duration": 79443792,
"eval_count": 13,
"eval_duration": 653000000,
"eval_duration": 504352750,
"response": "",
"thinking": null,
"context": null

178
tests/integration/recordings/responses/6f96090aa955.json
View file

@ -1,7 +1,7 @@
{
"request": {
"method": "POST",
"url": "http://localhost:11434/v1/v1/chat/completions",
"url": "http://0.0.0.0:11434/v1/v1/chat/completions",
"headers": {},
"body": {
"model": "llama3.2:3b-instruct-fp16",
@ -21,7 +21,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -36,7 +36,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -47,7 +47,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -62,7 +62,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -73,11 +73,11 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
"content": " Welcome",
"content": " It",
"function_call": null,
"refusal": null,
"role": "assistant",
@ -88,7 +88,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -99,7 +99,59 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
"content": "'s",
"function_call": null,
"refusal": null,
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
"system_fingerprint": "fp_ollama",
"usage": null
}
},
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-698",
"choices": [
{
"delta": {
"content": " nice",
"function_call": null,
"refusal": null,
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
"system_fingerprint": "fp_ollama",
"usage": null
}
},
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -114,7 +166,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -125,11 +177,11 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
"content": " our",
"content": " meet",
"function_call": null,
"refusal": null,
"role": "assistant",
@ -140,7 +192,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -151,11 +203,11 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
"content": " conversation",
"content": " you",
"function_call": null,
"refusal": null,
"role": "assistant",
@ -166,7 +218,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -177,7 +229,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -192,7 +244,7 @@
"logprobs": null
}
],
"created": 1754081849,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -203,7 +255,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -218,7 +270,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -229,7 +281,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -244,7 +296,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -255,7 +307,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -270,7 +322,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -281,7 +333,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -296,7 +348,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -307,7 +359,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -322,7 +374,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -333,7 +385,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -348,7 +400,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -359,7 +411,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -374,7 +426,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -385,7 +437,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -400,7 +452,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -411,33 +463,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"choices": [
{
"delta": {
"content": ",",
"function_call": null,
"refusal": null,
"role": "assistant",
"tool_calls": null
},
"finish_reason": null,
"index": 0,
"logprobs": null
}
],
"created": 1754081850,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
"system_fingerprint": "fp_ollama",
"usage": null
}
},
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -452,7 +478,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -463,7 +489,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -478,7 +504,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921359,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -489,7 +515,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -504,7 +530,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -515,7 +541,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -530,7 +556,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -541,7 +567,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -556,7 +582,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -567,7 +593,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -582,7 +608,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -593,7 +619,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -608,7 +634,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,
@ -619,7 +645,7 @@
{
"__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
"__data__": {
"id": "chatcmpl-333",
"id": "chatcmpl-698",
"choices": [
{
"delta": {
@ -634,7 +660,7 @@
"logprobs": null
}
],
"created": 1754081850,
"created": 1756921360,
"model": "llama3.2:3b-instruct-fp16",
"object": "chat.completion.chunk",
"service_tier": null,

46
tests/integration/recordings/responses/6fbea1abca7c.json
View file

@ -22,7 +22,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.337763Z",
"created_at": "2025-09-03T17:38:01.89965Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -40,7 +40,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.394358Z",
"created_at": "2025-09-03T17:38:01.941253Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -58,7 +58,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.451349Z",
"created_at": "2025-09-03T17:38:01.982621Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -76,7 +76,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.504443Z",
"created_at": "2025-09-03T17:38:02.024144Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -94,7 +94,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.555779Z",
"created_at": "2025-09-03T17:38:02.065495Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -112,7 +112,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.607807Z",
"created_at": "2025-09-03T17:38:02.107529Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -130,7 +130,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.660627Z",
"created_at": "2025-09-03T17:38:02.149217Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -148,7 +148,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.711562Z",
"created_at": "2025-09-03T17:38:02.190357Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -166,7 +166,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.761822Z",
"created_at": "2025-09-03T17:38:02.231501Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -184,7 +184,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.81712Z",
"created_at": "2025-09-03T17:38:02.272546Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -202,7 +202,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.868755Z",
"created_at": "2025-09-03T17:38:02.313561Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -220,7 +220,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.921049Z",
"created_at": "2025-09-03T17:38:02.354563Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -238,7 +238,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:20.973584Z",
"created_at": "2025-09-03T17:38:02.395585Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -256,7 +256,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.030707Z",
"created_at": "2025-09-03T17:38:02.436854Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -274,7 +274,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.082015Z",
"created_at": "2025-09-03T17:38:02.47814Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -292,7 +292,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.132945Z",
"created_at": "2025-09-03T17:38:02.519661Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -310,7 +310,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.187452Z",
"created_at": "2025-09-03T17:38:02.561119Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -328,7 +328,7 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.239827Z",
"created_at": "2025-09-03T17:38:02.602821Z",
"done": false,
"done_reason": null,
"total_duration": null,
@ -346,15 +346,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-01T23:14:21.294154Z",
"created_at": "2025-09-03T17:38:02.644633Z",
"done": true,
"done_reason": "stop",
"total_duration": 1929211666,
"load_duration": 61298666,
"total_duration": 1375629459,
"load_duration": 94090250,
"prompt_eval_count": 386,
"prompt_eval_duration": 908000000,
"prompt_eval_duration": 535119167,
"eval_count": 19,
"eval_duration": 959000000,
"eval_duration": 745684041,
"response": "",
"thinking": null,
"context": null

9570
tests/integration/recordings/responses/6fe1d4fedf12.json
View file

File diff suppressed because it is too large Load diff

10
tests/integration/recordings/responses/70adef2c30c4.json
View file

@ -20,15 +20,15 @@
"__type__": "ollama._types.GenerateResponse",
"__data__": {
"model": "llama3.2:3b-instruct-fp16",
"created_at": "2025-08-04T22:55:55.720345Z",
"created_at": "2025-09-03T17:42:17.227488Z",
"done": true,
"done_reason": "stop",
"total_duration": 3865701084,
"load_duration": 52435459,
"total_duration": 3003964916,
"load_duration": 111221916,
"prompt_eval_count": 30,
"prompt_eval_duration": 99000000,
"prompt_eval_duration": 72578583,
"eval_count": 70,
"eval_duration": 3712000000,
"eval_duration": 2819555375,
"response": "The answer is Saturn! Saturn's ring system is one of the most iconic and well-known in our solar system. The rings are made up of ice particles, rock debris, and dust that orbit around the planet due to its gravitational pull.\n\nWould you like to know more about Saturn's rings or is there something else I can help you with?",
"thinking": null,
"context": null

422
tests/integration/recordings/responses/72c1126ff2f9.json Normal file
View file

@ -0,0 +1,422 @@
{
"request": {
"method": "POST",
"url": "http://0.0.0.0:11434/v1/v1/embeddings",
"headers": {},
"body": {
"model": "all-minilm:l6-v2",
"input": [
"artificial intelligence"
],
"encoding_format": "float"
},
"endpoint": "/v1/embeddings",
"model": "all-minilm:l6-v2"
},
"response": {
"body": {
"__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
"__data__": {
"data": [
{
"embedding": [
-0.024330618,
0.016706783,
0.037677176,
-0.00915746,
-0.030534461,
-0.017140884,
0.074272,
0.0456916,
-0.009377196,
0.009883053,
-0.0056895507,
0.007668296,
0.039537333,
0.015226257,
-0.083189555,
0.019439526,
-0.022046678,
-0.033254813,
-0.18105465,
-0.13025087,
-0.0022671346,
0.013451522,
-0.024325468,
-0.0370128,
0.0020083552,
0.08566712,
0.0047639925,
-0.0033431018,
-0.006082307,
-0.11575565,
0.06682902,
-0.018777572,
0.08786827,
-0.0074177794,
-0.093573004,
0.06146399,
-0.08110609,
0.012222862,
0.03971064,
-0.0026197461,
-0.04657111,
-0.08183902,
0.03959615,
0.015451151,
0.04370617,
0.103643835,
-0.058421485,
0.036699355,
-0.052699573,
0.040590122,
-0.12578927,
0.006500531,
-0.03583627,
-0.010050973,
-0.023851713,
0.045972254,
0.014605586,
0.019414552,
0.028465148,
-0.055030964,
0.024210233,
-0.052867457,
0.015230711,
-0.0043921247,
0.092372045,
0.033849865,
-0.04737281,
0.03204496,
0.001322036,
-0.051211488,
0.025862284,
0.08155327,
0.04092595,
0.019154705,
0.056453932,
-0.052758913,
0.030533386,
-0.01663434,
0.07877244,
-0.054262977,
-0.042149354,
-0.045443602,
-0.052689902,
0.11225497,
0.01989102,
-0.042375352,
-0.01168115,
0.024315914,
0.01915792,
-0.016550383,
-0.01030883,
-0.08545277,
0.023834355,
-0.042181373,
-0.02503509,
0.062114798,
-0.0045557353,
-0.15369569,
0.001106691,
0.19423288,
-0.0338511,
0.026152972,
-0.02032091,
0.0012884078,
-0.0010269672,
-0.02411262,
0.017495485,
-0.009808713,
0.07037937,
-0.13769862,
-0.11118059,
-0.01736481,
0.06603106,
-0.05188892,
0.0019610007,
0.014606686,
0.060775463,
0.096280165,
0.013551965,
0.019343173,
-0.00010512453,
-0.026652312,
-0.009341819,
0.07083247,
-0.0034617546,
-0.062412772,
-0.044611085,
-8.796679e-34,
-0.111884,
-0.04256611,
0.027425196,
0.06574074,
0.002830377,
-0.044104468,
0.005238822,
-0.036899913,
-0.015583552,
0.0206543,
-0.059225976,
0.007236511,
-0.028716031,
0.040467348,
0.13387093,
0.006795838,
-0.01636956,
0.082198486,
-0.02261007,
-0.03641293,
0.06524453,
0.021011814,
-0.005472363,
-0.038433436,
0.001462021,
0.0073671984,
0.016773427,
-0.062663026,
0.035388503,
-0.014395795,
0.027888605,
0.0837546,
-0.027772024,
-0.0036210797,
0.03903557,
-0.026879627,
-0.018737236,
0.019059159,
0.06522148,
0.0070414003,
0.004749159,
-0.0030224407,
0.040062208,
0.028016094,
-0.004660955,
0.012264517,
0.08708117,
-0.0070171114,
-0.03749808,
0.011326775,
0.015419708,
0.013775354,
0.017958472,
-0.009817919,
0.09011542,
0.05170552,
-0.034259036,
0.0043903207,
-0.01884889,
-0.031481344,
0.08216297,
0.016875258,
-0.022163702,
0.06844141,
0.01581623,
0.020322658,
0.0063856863,
0.016461994,
0.12718283,
0.014996434,
-0.010813858,
0.0017669421,
0.03166716,
-0.044353984,
-0.05225622,
0.022843942,
0.050988898,
-0.018916955,
0.0027930918,
-0.033645593,
-0.13571611,
-0.027015164,
-0.035672266,
-0.033537813,
0.047864296,
-0.0054381513,
0.021346755,
-0.040034927,
0.019374551,
0.012011466,
-0.04336231,
0.00054701004,
0.034879614,
0.017960642,
-0.062501945,
8.224154e-34,
-0.09450138,
0.013776636,
-0.025351105,
0.098992504,
0.045503527,
-0.02053458,
-0.029694881,
-0.059200566,
0.042453792,
0.0844487,
-0.043211546,
-0.0077362363,
0.049354795,
0.04203366,
-0.036539596,
0.014424774,
0.040357023,
-0.058971472,
0.010022987,
0.059877146,
-0.02790864,
0.034927685,
-0.087597504,
-0.060616262,
-0.0048867166,
0.08776906,
-0.0053599468,
-0.021816833,
-0.048162397,
0.046919785,
0.0083988905,
-0.0517289,
-0.020422187,
0.08581073,
-0.022597926,
0.034425046,
-0.014506674,
0.0031332907,
-0.04651877,
0.030281488,
0.039713897,
0.02969227,
-0.09310218,
0.051527865,
0.007809,
-0.05700871,
-0.041792583,
0.08987064,
-0.00813404,
-0.04082285,
-0.053487595,
-0.034378976,
-0.045253906,
-0.09715307,
-0.058194414,
0.06093547,
-0.009079956,
0.006918499,
0.012345728,
0.062036473,
-0.0060238577,
-0.0864295,
0.05872831,
0.053304974,
-0.05352623,
0.039521407,
-0.04498403,
0.0727911,
-0.039616212,
-0.05134442,
0.10334881,
0.02176773,
0.00016648973,
0.009423309,
0.022016358,
-0.006902813,
-0.128883,
-0.009864072,
-0.036396757,
-0.042481646,
0.004420737,
-0.047660243,
0.0065179355,
0.102602735,
-0.053166825,
0.07328581,
0.015810944,
-0.029149039,
0.025130944,
-0.063055776,
-0.043462534,
0.06719971,
0.014921177,
-0.0010985207,
-0.09869465,
-1.4682753e-08,
0.004611013,
-0.06715223,
0.07644809,
-0.019802453,
0.06737909,
0.044783685,
-0.050963327,
-0.0077186874,
-0.029319718,
0.028867716,
0.018877175,
-0.024279349,
0.04412064,
0.04416273,
0.03432814,
0.046517964,
0.02158077,
-0.001748483,
-0.0029956794,
0.014355785,
0.12525895,
0.03431845,
-0.014617591,
0.039184693,
-0.0023036227,
-0.014352919,
0.01010173,
0.02430961,
-0.041730728,
0.08832413,
-0.031459343,
0.030073628,
-0.0029376182,
0.0049478672,
0.09588392,
0.09396655,
0.01412568,
-0.077148266,
-0.039246846,
-0.01064901,
-0.008556093,
0.06409403,
-0.033037152,
-0.03049978,
0.0945846,
-0.008954658,
-0.029921891,
-0.132985,
0.059934624,
-0.011668423,
0.0071737366,
0.035627652,
0.0041028745,
0.056198087,
0.07656151,
-0.010067092,
0.05678312,
0.023536043,
-0.063770495,
0.08934554,
0.043756966,
0.04337246,
0.046287052,
-0.07039028
],
"index": 0,
"object": "embedding"
}
],
"model": "all-minilm:l6-v2",
"object": "list",
"usage": {
"prompt_tokens": 2,
"total_tokens": 2
}
}
},
"is_streaming": false
}
}

Some files were not shown because too many files have changed in this diff Show more