mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-10-24 08:47:26 +00:00
refactor(agent): drop AgentToolGroup for responses tools
This commit is contained in:
Ashwin Bharambe
parent
c56b2deb7d
commit
ce44b9d6f6
12 changed files with 4051 additions and 4225 deletions
424
docs/static/stainless-llama-stack-spec.yaml
vendored
424
docs/static/stainless-llama-stack-spec.yaml
vendored
|
|
@ -11663,8 +11663,16 @@ components:
|
|||
AgentConfig:
|
||||
type: object
|
||||
properties:
|
||||
sampling_params:
|
||||
$ref: '#/components/schemas/SamplingParams'
|
||||
max_output_tokens:
|
||||
type: integer
|
||||
temperature:
|
||||
type: number
|
||||
top_p:
|
||||
type: number
|
||||
stop:
|
||||
type: array
|
||||
items:
|
||||
type: string
|
||||
input_shields:
|
||||
type: array
|
||||
items:
|
||||
|
|
@ -11673,36 +11681,16 @@ components:
|
|||
type: array
|
||||
items:
|
||||
type: string
|
||||
toolgroups:
|
||||
tools:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/AgentTool'
|
||||
$ref: '#/components/schemas/OpenAIResponseInputTool'
|
||||
client_tools:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/ToolDef'
|
||||
tool_choice:
|
||||
type: string
|
||||
enum:
|
||||
- auto
|
||||
- required
|
||||
- none
|
||||
title: ToolChoice
|
||||
description: >-
|
||||
Whether tool use is required or automatic. This is a hint to the model
|
||||
which may not be followed. It depends on the Instruction Following capabilities
|
||||
of the model.
|
||||
deprecated: true
|
||||
tool_prompt_format:
|
||||
type: string
|
||||
enum:
|
||||
- json
|
||||
- function_tag
|
||||
- python_list
|
||||
title: ToolPromptFormat
|
||||
description: >-
|
||||
Prompt format for calling custom / zero shot tools.
|
||||
deprecated: true
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/OpenAIResponseInputTool'
|
||||
- $ref: '#/components/schemas/ToolDef'
|
||||
tool_config:
|
||||
$ref: '#/components/schemas/ToolConfig'
|
||||
max_infer_iters:
|
||||
|
|
@ -11725,7 +11713,7 @@ components:
|
|||
description: >-
|
||||
Optional flag indicating whether session data has to be persisted
|
||||
response_format:
|
||||
$ref: '#/components/schemas/ResponseFormat'
|
||||
$ref: '#/components/schemas/OpenAIResponseFormatParam'
|
||||
description: Optional response format configuration
|
||||
additionalProperties: false
|
||||
required:
|
||||
|
|
@ -11733,157 +11721,6 @@ components:
|
|||
- instructions
|
||||
title: AgentConfig
|
||||
description: Configuration for an agent.
|
||||
AgentTool:
|
||||
oneOf:
|
||||
- type: string
|
||||
- type: object
|
||||
properties:
|
||||
name:
|
||||
type: string
|
||||
args:
|
||||
type: object
|
||||
additionalProperties:
|
||||
oneOf:
|
||||
- type: 'null'
|
||||
- type: boolean
|
||||
- type: number
|
||||
- type: string
|
||||
- type: array
|
||||
- type: object
|
||||
additionalProperties: false
|
||||
required:
|
||||
- name
|
||||
- args
|
||||
title: AgentToolGroupWithArgs
|
||||
GrammarResponseFormat:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
enum:
|
||||
- json_schema
|
||||
- grammar
|
||||
description: >-
|
||||
Must be "grammar" to identify this format type
|
||||
const: grammar
|
||||
default: grammar
|
||||
bnf:
|
||||
type: object
|
||||
additionalProperties:
|
||||
oneOf:
|
||||
- type: 'null'
|
||||
- type: boolean
|
||||
- type: number
|
||||
- type: string
|
||||
- type: array
|
||||
- type: object
|
||||
description: >-
|
||||
The BNF grammar specification the response should conform to
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
- bnf
|
||||
title: GrammarResponseFormat
|
||||
description: >-
|
||||
Configuration for grammar-guided response generation.
|
||||
GreedySamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: greedy
|
||||
default: greedy
|
||||
description: >-
|
||||
Must be "greedy" to identify this sampling strategy
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
title: GreedySamplingStrategy
|
||||
description: >-
|
||||
Greedy sampling strategy that selects the highest probability token at each
|
||||
step.
|
||||
JsonSchemaResponseFormat:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
enum:
|
||||
- json_schema
|
||||
- grammar
|
||||
description: >-
|
||||
Must be "json_schema" to identify this format type
|
||||
const: json_schema
|
||||
default: json_schema
|
||||
json_schema:
|
||||
type: object
|
||||
additionalProperties:
|
||||
oneOf:
|
||||
- type: 'null'
|
||||
- type: boolean
|
||||
- type: number
|
||||
- type: string
|
||||
- type: array
|
||||
- type: object
|
||||
description: >-
|
||||
The JSON schema the response should conform to. In a Python SDK, this
|
||||
is often a `pydantic` model.
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
- json_schema
|
||||
title: JsonSchemaResponseFormat
|
||||
description: >-
|
||||
Configuration for JSON schema-guided response generation.
|
||||
ResponseFormat:
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/JsonSchemaResponseFormat'
|
||||
- $ref: '#/components/schemas/GrammarResponseFormat'
|
||||
discriminator:
|
||||
propertyName: type
|
||||
mapping:
|
||||
json_schema: '#/components/schemas/JsonSchemaResponseFormat'
|
||||
grammar: '#/components/schemas/GrammarResponseFormat'
|
||||
SamplingParams:
|
||||
type: object
|
||||
properties:
|
||||
strategy:
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/GreedySamplingStrategy'
|
||||
- $ref: '#/components/schemas/TopPSamplingStrategy'
|
||||
- $ref: '#/components/schemas/TopKSamplingStrategy'
|
||||
discriminator:
|
||||
propertyName: type
|
||||
mapping:
|
||||
greedy: '#/components/schemas/GreedySamplingStrategy'
|
||||
top_p: '#/components/schemas/TopPSamplingStrategy'
|
||||
top_k: '#/components/schemas/TopKSamplingStrategy'
|
||||
description: The sampling strategy.
|
||||
max_tokens:
|
||||
type: integer
|
||||
default: 0
|
||||
description: >-
|
||||
The maximum number of tokens that can be generated in the completion.
|
||||
The token count of your prompt plus max_tokens cannot exceed the model's
|
||||
context length.
|
||||
repetition_penalty:
|
||||
type: number
|
||||
default: 1.0
|
||||
description: >-
|
||||
Number between -2.0 and 2.0. Positive values penalize new tokens based
|
||||
on whether they appear in the text so far, increasing the model's likelihood
|
||||
to talk about new topics.
|
||||
stop:
|
||||
type: array
|
||||
items:
|
||||
type: string
|
||||
description: >-
|
||||
Up to 4 sequences where the API will stop generating further tokens. The
|
||||
returned text will not contain the stop sequence.
|
||||
additionalProperties: false
|
||||
required:
|
||||
- strategy
|
||||
title: SamplingParams
|
||||
description: Sampling parameters.
|
||||
ToolConfig:
|
||||
type: object
|
||||
properties:
|
||||
|
|
@ -11932,51 +11769,6 @@ components:
|
|||
additionalProperties: false
|
||||
title: ToolConfig
|
||||
description: Configuration for tool use.
|
||||
TopKSamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: top_k
|
||||
default: top_k
|
||||
description: >-
|
||||
Must be "top_k" to identify this sampling strategy
|
||||
top_k:
|
||||
type: integer
|
||||
description: >-
|
||||
Number of top tokens to consider for sampling. Must be at least 1
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
- top_k
|
||||
title: TopKSamplingStrategy
|
||||
description: >-
|
||||
Top-k sampling strategy that restricts sampling to the k most likely tokens.
|
||||
TopPSamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: top_p
|
||||
default: top_p
|
||||
description: >-
|
||||
Must be "top_p" to identify this sampling strategy
|
||||
temperature:
|
||||
type: number
|
||||
description: >-
|
||||
Controls randomness in sampling. Higher values increase randomness
|
||||
top_p:
|
||||
type: number
|
||||
default: 0.95
|
||||
description: >-
|
||||
Cumulative probability threshold for nucleus sampling. Defaults to 0.95
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
title: TopPSamplingStrategy
|
||||
description: >-
|
||||
Top-p (nucleus) sampling strategy that samples from the smallest set of tokens
|
||||
with cumulative probability >= p.
|
||||
CreateAgentRequest:
|
||||
type: object
|
||||
properties:
|
||||
|
|
@ -12072,8 +11864,10 @@ components:
|
|||
const: inference
|
||||
default: inference
|
||||
model_response:
|
||||
$ref: '#/components/schemas/CompletionMessage'
|
||||
$ref: '#/components/schemas/OpenAIAssistantMessageParam'
|
||||
description: The response from the LLM.
|
||||
finish_reason:
|
||||
type: string
|
||||
additionalProperties: false
|
||||
required:
|
||||
- turn_id
|
||||
|
|
@ -12115,7 +11909,11 @@ components:
|
|||
description: >-
|
||||
The IDs of the vector databases to retrieve context from.
|
||||
inserted_context:
|
||||
$ref: '#/components/schemas/InterleavedContent'
|
||||
oneOf:
|
||||
- type: string
|
||||
- type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/OpenAIChatCompletionContentPartParam'
|
||||
description: >-
|
||||
The context retrieved from the vector databases.
|
||||
additionalProperties: false
|
||||
|
|
@ -12226,12 +12024,12 @@ components:
|
|||
tool_calls:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/ToolCall'
|
||||
$ref: '#/components/schemas/OpenAIChatCompletionToolCall'
|
||||
description: The tool calls to execute.
|
||||
tool_responses:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/ToolResponse'
|
||||
$ref: '#/components/schemas/OpenAIToolMessageParam'
|
||||
description: The tool responses from the tool calls.
|
||||
additionalProperties: false
|
||||
required:
|
||||
|
|
@ -12242,46 +12040,6 @@ components:
|
|||
- tool_responses
|
||||
title: ToolExecutionStep
|
||||
description: A tool execution step in an agent turn.
|
||||
ToolResponse:
|
||||
type: object
|
||||
properties:
|
||||
call_id:
|
||||
type: string
|
||||
description: >-
|
||||
Unique identifier for the tool call this response is for
|
||||
tool_name:
|
||||
oneOf:
|
||||
- type: string
|
||||
enum:
|
||||
- brave_search
|
||||
- wolfram_alpha
|
||||
- photogen
|
||||
- code_interpreter
|
||||
title: BuiltinTool
|
||||
- type: string
|
||||
description: Name of the tool that was invoked
|
||||
content:
|
||||
$ref: '#/components/schemas/InterleavedContent'
|
||||
description: The response content from the tool
|
||||
metadata:
|
||||
type: object
|
||||
additionalProperties:
|
||||
oneOf:
|
||||
- type: 'null'
|
||||
- type: boolean
|
||||
- type: number
|
||||
- type: string
|
||||
- type: array
|
||||
- type: object
|
||||
description: >-
|
||||
(Optional) Additional metadata about the tool response
|
||||
additionalProperties: false
|
||||
required:
|
||||
- call_id
|
||||
- tool_name
|
||||
- content
|
||||
title: ToolResponse
|
||||
description: Response from a tool invocation.
|
||||
Turn:
|
||||
type: object
|
||||
properties:
|
||||
|
|
@ -12296,9 +12054,7 @@ components:
|
|||
input_messages:
|
||||
type: array
|
||||
items:
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/UserMessage'
|
||||
- $ref: '#/components/schemas/ToolResponseMessage'
|
||||
$ref: '#/components/schemas/OpenAIMessageParam'
|
||||
description: >-
|
||||
List of messages that initiated this turn
|
||||
steps:
|
||||
|
|
@ -12319,9 +12075,11 @@ components:
|
|||
description: >-
|
||||
Ordered list of processing steps executed during this turn
|
||||
output_message:
|
||||
$ref: '#/components/schemas/CompletionMessage'
|
||||
$ref: '#/components/schemas/OpenAIAssistantMessageParam'
|
||||
description: >-
|
||||
The model's generated response containing content and metadata
|
||||
finish_reason:
|
||||
type: string
|
||||
output_attachments:
|
||||
type: array
|
||||
items:
|
||||
|
|
@ -12330,10 +12088,9 @@ components:
|
|||
content:
|
||||
oneOf:
|
||||
- type: string
|
||||
- $ref: '#/components/schemas/InterleavedContentItem'
|
||||
- type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/InterleavedContentItem'
|
||||
$ref: '#/components/schemas/OpenAIChatCompletionContentPartParam'
|
||||
- $ref: '#/components/schemas/URL'
|
||||
description: The content of the attachment.
|
||||
mime_type:
|
||||
|
|
@ -12373,9 +12130,7 @@ components:
|
|||
messages:
|
||||
type: array
|
||||
items:
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/UserMessage'
|
||||
- $ref: '#/components/schemas/ToolResponseMessage'
|
||||
$ref: '#/components/schemas/OpenAIMessageParam'
|
||||
description: List of messages to start the turn with.
|
||||
stream:
|
||||
type: boolean
|
||||
|
|
@ -12390,10 +12145,9 @@ components:
|
|||
content:
|
||||
oneOf:
|
||||
- type: string
|
||||
- $ref: '#/components/schemas/InterleavedContentItem'
|
||||
- type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/InterleavedContentItem'
|
||||
$ref: '#/components/schemas/OpenAIChatCompletionContentPartParam'
|
||||
- $ref: '#/components/schemas/URL'
|
||||
description: The content of the document.
|
||||
mime_type:
|
||||
|
|
@ -12407,13 +12161,13 @@ components:
|
|||
description: A document to be used by an agent.
|
||||
description: >-
|
||||
(Optional) List of documents to create the turn with.
|
||||
toolgroups:
|
||||
tools:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/AgentTool'
|
||||
$ref: '#/components/schemas/OpenAIResponseInputTool'
|
||||
description: >-
|
||||
(Optional) List of toolgroups to create the turn with, will be used in
|
||||
addition to the agent's config toolgroups for the request.
|
||||
(Optional) List of tools to create the turn with, will be used in addition
|
||||
to the agent's config tools for the request.
|
||||
tool_config:
|
||||
$ref: '#/components/schemas/ToolConfig'
|
||||
description: >-
|
||||
|
|
@ -12764,7 +12518,7 @@ components:
|
|||
tool_responses:
|
||||
type: array
|
||||
items:
|
||||
$ref: '#/components/schemas/ToolResponse'
|
||||
$ref: '#/components/schemas/OpenAIToolMessageParam'
|
||||
description: >-
|
||||
The tool call responses to resume the turn with.
|
||||
stream:
|
||||
|
|
@ -12955,6 +12709,22 @@ components:
|
|||
title: BenchmarkConfig
|
||||
description: >-
|
||||
A benchmark configuration for evaluation.
|
||||
GreedySamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: greedy
|
||||
default: greedy
|
||||
description: >-
|
||||
Must be "greedy" to identify this sampling strategy
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
title: GreedySamplingStrategy
|
||||
description: >-
|
||||
Greedy sampling strategy that selects the highest probability token at each
|
||||
step.
|
||||
ModelCandidate:
|
||||
type: object
|
||||
properties:
|
||||
|
|
@ -12980,6 +12750,92 @@ components:
|
|||
- sampling_params
|
||||
title: ModelCandidate
|
||||
description: A model candidate for evaluation.
|
||||
SamplingParams:
|
||||
type: object
|
||||
properties:
|
||||
strategy:
|
||||
oneOf:
|
||||
- $ref: '#/components/schemas/GreedySamplingStrategy'
|
||||
- $ref: '#/components/schemas/TopPSamplingStrategy'
|
||||
- $ref: '#/components/schemas/TopKSamplingStrategy'
|
||||
discriminator:
|
||||
propertyName: type
|
||||
mapping:
|
||||
greedy: '#/components/schemas/GreedySamplingStrategy'
|
||||
top_p: '#/components/schemas/TopPSamplingStrategy'
|
||||
top_k: '#/components/schemas/TopKSamplingStrategy'
|
||||
description: The sampling strategy.
|
||||
max_tokens:
|
||||
type: integer
|
||||
default: 0
|
||||
description: >-
|
||||
The maximum number of tokens that can be generated in the completion.
|
||||
The token count of your prompt plus max_tokens cannot exceed the model's
|
||||
context length.
|
||||
repetition_penalty:
|
||||
type: number
|
||||
default: 1.0
|
||||
description: >-
|
||||
Number between -2.0 and 2.0. Positive values penalize new tokens based
|
||||
on whether they appear in the text so far, increasing the model's likelihood
|
||||
to talk about new topics.
|
||||
stop:
|
||||
type: array
|
||||
items:
|
||||
type: string
|
||||
description: >-
|
||||
Up to 4 sequences where the API will stop generating further tokens. The
|
||||
returned text will not contain the stop sequence.
|
||||
additionalProperties: false
|
||||
required:
|
||||
- strategy
|
||||
title: SamplingParams
|
||||
description: Sampling parameters.
|
||||
TopKSamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: top_k
|
||||
default: top_k
|
||||
description: >-
|
||||
Must be "top_k" to identify this sampling strategy
|
||||
top_k:
|
||||
type: integer
|
||||
description: >-
|
||||
Number of top tokens to consider for sampling. Must be at least 1
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
- top_k
|
||||
title: TopKSamplingStrategy
|
||||
description: >-
|
||||
Top-k sampling strategy that restricts sampling to the k most likely tokens.
|
||||
TopPSamplingStrategy:
|
||||
type: object
|
||||
properties:
|
||||
type:
|
||||
type: string
|
||||
const: top_p
|
||||
default: top_p
|
||||
description: >-
|
||||
Must be "top_p" to identify this sampling strategy
|
||||
temperature:
|
||||
type: number
|
||||
description: >-
|
||||
Controls randomness in sampling. Higher values increase randomness
|
||||
top_p:
|
||||
type: number
|
||||
default: 0.95
|
||||
description: >-
|
||||
Cumulative probability threshold for nucleus sampling. Defaults to 0.95
|
||||
additionalProperties: false
|
||||
required:
|
||||
- type
|
||||
title: TopPSamplingStrategy
|
||||
description: >-
|
||||
Top-p (nucleus) sampling strategy that samples from the smallest set of tokens
|
||||
with cumulative probability >= p.
|
||||
EvaluateRowsRequest:
|
||||
type: object
|
||||
properties:
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue