phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-03 18:00:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge 6fb329d613 into sapling-pr-archive-ehhuang

Browse source
This commit is contained in:
ehhuang 2025-11-05 14:11:07 -08:00 • committed by GitHub
commit 913d9717b6
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 212 additions and 230 deletions
Download patch file Download diff file Expand all files Collapse all files

140
client-sdks/stainless/openapi.yml
View file

@ -9976,6 +9976,70 @@ components:
- metadata
title: VectorStoreObject
description: OpenAI Vector Store object.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreRequestWithExtraBody":
type: object
properties:
@ -10001,15 +10065,7 @@ components:
description: >-
(Optional) Expiration policy for the vector store
chunking_strategy:
type: object
additionalProperties:
oneOf:
- type: 'null'
- type: boolean
- type: number
- type: string
- type: array
- type: object
$ref: '#/components/schemas/VectorStoreChunkingStrategy'
description: >-
(Optional) Strategy for splitting files into chunks
metadata:
@ -10085,70 +10141,6 @@ components:
- deleted
title: VectorStoreDeleteResponse
description: Response from deleting a vector store.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreFileBatchRequestWithExtraBody":
type: object
properties:
@ -10606,6 +10598,8 @@ components:
description: >-
Object type identifier for the search results page
search_query:
type: array
items:
type: string
description: >-
The original search query that was executed

140
docs/static/llama-stack-spec.yaml vendored
View file

@ -9260,6 +9260,70 @@ components:
- metadata
title: VectorStoreObject
description: OpenAI Vector Store object.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreRequestWithExtraBody":
type: object
properties:
@ -9285,15 +9349,7 @@ components:
description: >-
(Optional) Expiration policy for the vector store
chunking_strategy:
type: object
additionalProperties:
oneOf:
- type: 'null'
- type: boolean
- type: number
- type: string
- type: array
- type: object
$ref: '#/components/schemas/VectorStoreChunkingStrategy'
description: >-
(Optional) Strategy for splitting files into chunks
metadata:
@ -9369,70 +9425,6 @@ components:
- deleted
title: VectorStoreDeleteResponse
description: Response from deleting a vector store.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreFileBatchRequestWithExtraBody":
type: object
properties:
@ -9890,6 +9882,8 @@ components:
description: >-
Object type identifier for the search results page
search_query:
type: array
items:
type: string
description: >-
The original search query that was executed

140
docs/static/stainless-llama-stack-spec.yaml vendored
View file

@ -9976,6 +9976,70 @@ components:
- metadata
title: VectorStoreObject
description: OpenAI Vector Store object.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreRequestWithExtraBody":
type: object
properties:
@ -10001,15 +10065,7 @@ components:
description: >-
(Optional) Expiration policy for the vector store
chunking_strategy:
type: object
additionalProperties:
oneOf:
- type: 'null'
- type: boolean
- type: number
- type: string
- type: array
- type: object
$ref: '#/components/schemas/VectorStoreChunkingStrategy'
description: >-
(Optional) Strategy for splitting files into chunks
metadata:
@ -10085,70 +10141,6 @@ components:
- deleted
title: VectorStoreDeleteResponse
description: Response from deleting a vector store.
VectorStoreChunkingStrategy:
oneOf:
- $ref: '#/components/schemas/VectorStoreChunkingStrategyAuto'
- $ref: '#/components/schemas/VectorStoreChunkingStrategyStatic'
discriminator:
propertyName: type
mapping:
auto: '#/components/schemas/VectorStoreChunkingStrategyAuto'
static: '#/components/schemas/VectorStoreChunkingStrategyStatic'
VectorStoreChunkingStrategyAuto:
type: object
properties:
type:
type: string
const: auto
default: auto
description: >-
Strategy type, always "auto" for automatic chunking
additionalProperties: false
required:
- type
title: VectorStoreChunkingStrategyAuto
description: >-
Automatic chunking strategy for vector store files.
VectorStoreChunkingStrategyStatic:
type: object
properties:
type:
type: string
const: static
default: static
description: >-
Strategy type, always "static" for static chunking
static:
$ref: '#/components/schemas/VectorStoreChunkingStrategyStaticConfig'
description: >-
Configuration parameters for the static chunking strategy
additionalProperties: false
required:
- type
- static
title: VectorStoreChunkingStrategyStatic
description: >-
Static chunking strategy with configurable parameters.
VectorStoreChunkingStrategyStaticConfig:
type: object
properties:
chunk_overlap_tokens:
type: integer
default: 400
description: >-
Number of tokens to overlap between adjacent chunks
max_chunk_size_tokens:
type: integer
default: 800
description: >-
Maximum number of tokens per chunk, must be between 100 and 4096
additionalProperties: false
required:
- chunk_overlap_tokens
- max_chunk_size_tokens
title: VectorStoreChunkingStrategyStaticConfig
description: >-
Configuration for static chunking strategy.
"OpenAICreateVectorStoreFileBatchRequestWithExtraBody":
type: object
properties:
@ -10606,6 +10598,8 @@ components:
description: >-
Object type identifier for the search results page
search_query:
type: array
items:
type: string
description: >-
The original search query that was executed

4
src/llama_stack/apis/vector_io/vector_io.py
View file

@ -260,7 +260,7 @@ class VectorStoreSearchResponsePage(BaseModel):
"""
object: str = "vector_store.search_results.page"
search_query: str
search_query: list[str]
data: list[VectorStoreSearchResponse]
has_more: bool = False
next_page: str | None = None
@ -478,7 +478,7 @@ class OpenAICreateVectorStoreRequestWithExtraBody(BaseModel, extra="allow"):
name: str | None = None
file_ids: list[str] | None = None
expires_after: dict[str, Any] | None = None
chunking_strategy: dict[str, Any] | None = None
chunking_strategy: VectorStoreChunkingStrategy | None = None
metadata: dict[str, Any] | None = None

8
src/llama_stack/providers/utils/memory/openai_vector_store_mixin.py
View file

@ -642,7 +642,7 @@ class OpenAIVectorStoreMixin(ABC):
break
return VectorStoreSearchResponsePage(
search_query=search_query,
search_query=query if isinstance(query, list) else [query],
data=data,
has_more=False, # For simplicity, we don't implement pagination here
next_page=None,
@ -652,7 +652,7 @@ class OpenAIVectorStoreMixin(ABC):
logger.error(f"Error searching vector store {vector_store_id}: {e}")
# Return empty results on error
return VectorStoreSearchResponsePage(
search_query=search_query,
search_query=query if isinstance(query, list) else [query],
data=[],
has_more=False,
next_page=None,
@ -891,8 +891,8 @@ class OpenAIVectorStoreMixin(ABC):
# Determine pagination info
has_more = len(file_objects) > limit
first_id = file_objects[0].id if file_objects else None
last_id = file_objects[-1].id if file_objects else None
first_id = limited_files[0].id if file_objects else None
last_id = limited_files[-1].id if file_objects else None
return VectorStoreListFilesResponse(
data=limited_files,

4
tests/integration/vector_io/test_openai_vector_stores.py
View file

@ -350,7 +350,7 @@ def test_openai_vector_store_search_empty(
assert search_response is not None
assert hasattr(search_response, "data")
assert len(search_response.data) == 0 # Empty store should return no results
assert search_response.search_query == "test query"
assert search_response.search_query == ["test query"]
assert search_response.has_more is False
@ -825,7 +825,7 @@ def test_openai_vector_store_list_files(
assert first_page.has_more
assert len(first_page.data) == 2
assert first_page.first_id == first_page.data[0].id
assert first_page.last_id != first_page.data[-1].id
assert first_page.last_id == first_page.data[-1].id
next_page = compat_client.vector_stores.files.list(
vector_store_id=vector_store.id, limit=2, after=first_page.data[-1].id