> ## Documentation Index
> Fetch the complete documentation index at: https://docs.fireworks.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Create a Message
> **Anthropic-compatible endpoint.**
Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless multi-turn conversations.
**Fireworks Quickstarts:**
- [Serverless Quickstart](/getting-started/quickstart)
- [Deployments Quickstart](/getting-started/ondemand-quickstart)
This endpoint provides an Anthropic-compatible Messages API surface on Fireworks.
For setup, supported features, and known differences, see [Anthropic compatibility](/tools-sdks/anthropic-compatibility).
## OpenAPI
````yaml post /v1/messages
openapi: 3.1.0
info:
title: Fireworks AI Anthropic Compatible Messages API
description: Anthropic-compatible Messages API endpoint for Fireworks AI inference
version: 1.0.0
servers:
- url: https://api.fireworks.ai/inference
security:
- BearerAuth: []
paths:
/v1/messages:
post:
summary: Create a Message
description: >-
**Anthropic-compatible endpoint.**
Send a structured list of input messages with text and/or image content,
and the model will generate the next message in the conversation.
The Messages API can be used for either single queries or stateless
multi-turn conversations.
**Fireworks Quickstarts:**
- [Serverless Quickstart](/getting-started/quickstart)
- [Deployments Quickstart](/getting-started/ondemand-quickstart)
operationId: messages_post
requestBody:
content:
application/json:
schema:
$ref: '#/components/schemas/AnthropicCreateMessageParams'
required: true
responses:
'200':
description: Message object.
content:
application/json:
schema:
$ref: '#/components/schemas/AnthropicMessage'
4XX:
description: >-
Error response.
See the [errors documentation](/guides/inference-error-codes) for
more details.
content:
application/json:
schema:
$ref: '#/components/schemas/AnthropicErrorResponse'
components:
schemas:
AnthropicCreateMessageParams:
additionalProperties: false
example:
max_tokens: 1024
messages:
- content: Hello, world
role: user
model: claude-opus-4-6
properties:
model:
$ref: '#/components/schemas/AnthropicModel'
messages:
description: >-
Input messages.
Models are trained to operate on alternating `user` and `assistant`
conversational turns. When creating a new `Message`, you specify the
prior conversational turns with the `messages` parameter, and the
model then generates the next `Message` in the conversation.
Consecutive `user` or `assistant` turns in your request will be
combined into a single turn.
Each input message must be an object with a `role` and `content`.
You can specify a single `user`-role message, or you can include
multiple `user` and `assistant` messages.
If the final message uses the `assistant` role, the response content
will continue immediately from the content in that message. This can
be used to constrain part of the model's response.
Example with a single `user` message:
```json
[{"role": "user", "content": "Hello"}]
```
Example with multiple conversational turns:
```json
[
{"role": "user", "content": "Hello there."},
{"role": "assistant", "content": "Hi, I'm here to help. How can I help you?"},
{"role": "user", "content": "Can you explain LLMs in plain English?"},
]
```
Example with a partially-filled response from the model:
```json
[
{"role": "user", "content": "What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun"},
{"role": "assistant", "content": "The best answer is ("},
]
```
Each input message `content` may be either a single `string` or an
array of content blocks, where each block has a specific `type`.
Using a `string` for `content` is shorthand for an array of one
content block of type `"text"`. The following input messages are
equivalent:
```json
{"role": "user", "content": "Hello"}
```
```json
{"role": "user", "content": [{"type": "text", "text": "Hello"}]}
```
See [input
examples](https://docs.claude.com/en/api/messages-examples).
Note that if you want to include a [system
prompt](/guides/querying-text-models), you can use the top-level
`system` parameter — there is no `"system"` role for input messages
in the Messages API.
There is a limit of 100,000 messages in a single request.
items:
$ref: '#/components/schemas/AnthropicInputMessage'
title: Messages
type: array
max_tokens:
description: >-
The maximum number of tokens to generate before stopping.
Note that models may stop _before_ reaching this maximum. This
parameter only specifies the absolute maximum number of tokens to
generate.
Different models have different maximum values for this parameter.
See [models](https://app.fireworks.ai/models) for details.
examples:
- 1024
minimum: 1
title: Max Tokens
type: integer
metadata:
$ref: '#/components/schemas/AnthropicMetadata'
description: An object describing metadata about the request.
output_config:
$ref: '#/components/schemas/AnthropicOutputConfig'
description: >-
Configuration options for the model's output, such as the output
format.
stop_sequences:
description: >-
Custom text sequences that will cause the model to stop generating.
Models will normally stop when they have naturally completed their
turn, which will result in a response `stop_reason` of `"end_turn"`.
If you want the model to stop generating when it encounters custom
strings of text, you can use the `stop_sequences` parameter. If the
model encounters one of the custom sequences, the response
`stop_reason` value will be `"stop_sequence"` and the response
`stop_sequence` value will contain the matched stop sequence.
items:
type: string
title: Stop Sequences
type: array
stream:
description: >-
Whether to incrementally stream the response using server-sent
events.
See [streaming](/guides/querying-text-models) for details.
title: Stream
type: boolean
system:
anyOf:
- type: string
x-stainless-skip:
- go
- cli
- items:
$ref: '#/components/schemas/AnthropicRequestTextBlock'
type: array
description: >-
System prompt.
A system prompt is a way of providing context and instructions to
the model, such as specifying a particular goal or role. See the
[guide to system prompts](/guides/querying-text-models).
examples:
- - text: Today's date is 2024-06-01.
type: text
- Today's date is 2023-01-01.
title: System
temperature:
description: >-
Amount of randomness injected into the response.
Defaults to `1.0`. Ranges from `0.0` to `1.0`. Use `temperature`
closer to `0.0` for analytical / multiple choice, and closer to
`1.0` for creative and generative tasks.
Note that even with `temperature` of `0.0`, the results will not be
fully deterministic.
examples:
- 1
maximum: 1
minimum: 0
title: Temperature
type: number
thinking:
$ref: '#/components/schemas/AnthropicThinkingConfigParam'
tool_choice:
$ref: '#/components/schemas/AnthropicToolChoice'
tools:
description: >-
Definitions of tools that the model may use.
If you include `tools` in your API request, the model may return
`tool_use` content blocks that represent the model's use of those
tools. You can then run those tools using the tool input generated
by the model and then optionally return results back to the model
using `tool_result` content blocks.
Each tool definition includes:
* `name`: Name of the tool.
* `description`: Optional, but strongly-recommended description of
the tool.
* `input_schema`: [JSON
schema](https://json-schema.org/draft/2020-12) for the tool `input`
shape that the model will produce in `tool_use` output content
blocks.
For example, if you defined `tools` as:
```json
[
{
"name": "get_stock_price",
"description": "Get the current stock price for a given ticker symbol.",
"input_schema": {
"type": "object",
"properties": {
"ticker": {
"type": "string",
"description": "The stock ticker symbol, e.g. AAPL for Apple Inc."
}
},
"required": ["ticker"]
}
}
]
```
And then asked the model "What's the S&P 500 at today?", the model
might produce `tool_use` content blocks in the response like this:
```json
[
{
"type": "tool_use",
"id": "toolu_01D7FLrfh4GYq7yT1ULFeyMV",
"name": "get_stock_price",
"input": { "ticker": "^GSPC" }
}
]
```
You might then run your `get_stock_price` tool with `{"ticker":
"^GSPC"}` as an input, and return the following back to the model in
a subsequent `user` message:
```json
[
{
"type": "tool_result",
"tool_use_id": "toolu_01D7FLrfh4GYq7yT1ULFeyMV",
"content": "259.75 USD"
}
]
```
Tools can be used for workflows that include running client-side
tools and functions, or more generally whenever you want the model
to produce a particular JSON structure of output.
See the [guide](/guides/function-calling) for more details.
examples:
- description: Get the current weather in a given location
input_schema:
properties:
location:
description: The city and state, e.g. San Francisco, CA
type: string
unit:
description: Unit for the output - one of (celsius, fahrenheit)
type: string
required:
- location
type: object
name: get_weather
items:
oneOf:
- $ref: '#/components/schemas/AnthropicTool'
title: Tools
type: array
top_k:
description: >-
Only sample from the top K options for each subsequent token.
Used to remove "long tail" low probability responses. [Learn more
technical details
here](https://towardsdatascience.com/how-to-sample-from-language-models-682bceb97277).
Recommended for advanced use cases only. You usually only need to
use `temperature`.
examples:
- 5
minimum: 0
title: Top K
type: integer
top_p:
description: >-
Use nucleus sampling.
In nucleus sampling, we compute the cumulative distribution over all
the options for each subsequent token in decreasing probability
order and cut it off once it reaches a particular probability
specified by `top_p`. You should either alter `temperature` or
`top_p`, but not both.
Recommended for advanced use cases only. You usually only need to
use `temperature`.
examples:
- 0.7
maximum: 1
minimum: 0
title: Top P
type: number
raw_output:
anyOf:
- type: boolean
- type: 'null'
title: Raw Output
description: Return raw output from the model.
default: false
required:
- model
- messages
title: CreateMessageParams
type: object
AnthropicMessage:
examples:
- content:
- citations: null
text: Hi! How can I help you today?
type: text
id: msg_013Zva2CMHLNnXjNJJKqJ2EF
model: claude-opus-4-6
role: assistant
stop_reason: end_turn
stop_sequence: null
type: message
properties:
id:
description: |-
Unique object identifier.
The format and length of IDs may change over time.
examples:
- msg_013Zva2CMHLNnXjNJJKqJ2EF
title: Id
type: string
type:
const: message
default: message
description: |-
Object type.
For Messages, this is always `"message"`.
title: Type
type: string
role:
const: assistant
default: assistant
description: |-
Conversational role of the generated message.
This will always be `"assistant"`.
title: Role
type: string
content:
description: >-
Content generated by the model.
This is an array of content blocks, each of which has a `type` that
determines its shape.
Example:
```json
[{"type": "text", "text": "Hi, I'm here to help."}]
```
If the request input `messages` ended with an `assistant` turn, then
the response `content` will continue directly from that last turn.
You can use this to constrain the model's output.
For example, if the input `messages` were:
```json
[
{"role": "user", "content": "What's the Greek name for Sun? (A) Sol (B) Helios (C) Sun"},
{"role": "assistant", "content": "The best answer is ("}
]
```
Then the response `content` might be:
```json
[{"type": "text", "text": "B)"}]
```
examples:
- - citations: null
text: Hi! How can I help you today?
type: text
items:
$ref: '#/components/schemas/AnthropicContentBlock'
title: Content
type: array
model:
$ref: '#/components/schemas/AnthropicModel'
stop_reason:
anyOf:
- $ref: '#/components/schemas/AnthropicStopReason'
- type: 'null'
description: >-
The reason that the model stopped.
This may be one the following values:
* `"end_turn"`: the model reached a natural stopping point
* `"max_tokens"`: the model exceeded the requested `max_tokens` or
the model's maximum
* `"stop_sequence"`: one of your provided custom `stop_sequences`
was generated
* `"tool_use"`: the model invoked one or more tools
* `"pause_turn"`: the model paused a long-running turn. You may
provide the response back as-is in a subsequent request to let the
model continue.
* `"refusal"`: when streaming classifiers intervene to handle
potential policy violations
In non-streaming mode this value is always non-null. In streaming
mode, it is null in the `message_start` event and non-null
otherwise.
title: Stop Reason
stop_sequence:
anyOf:
- type: string
- type: 'null'
default: null
description: >-
Which custom stop sequence was generated, if any.
This value will be a non-null string if one of your custom stop
sequences was generated.
title: Stop Sequence
raw_output:
anyOf:
- $ref: '#/components/schemas/AnthropicRawOutput'
- type: 'null'
default: null
required:
- id
- type
- role
- content
- model
- stop_reason
- stop_sequence
title: Message
type: object
x-stainless-python-custom-imports:
- from .content_block import ContentBlock as ContentBlock
AnthropicErrorResponse:
properties:
error:
discriminator:
mapping:
api_error:
$ref: '#/components/schemas/AnthropicAPIError'
authentication_error:
$ref: '#/components/schemas/AnthropicAuthenticationError'
billing_error:
$ref: '#/components/schemas/AnthropicBillingError'
invalid_request_error:
$ref: '#/components/schemas/AnthropicInvalidRequestError'
not_found_error:
$ref: '#/components/schemas/AnthropicNotFoundError'
overloaded_error:
$ref: '#/components/schemas/AnthropicOverloadedError'
permission_error:
$ref: '#/components/schemas/AnthropicPermissionError'
rate_limit_error:
$ref: '#/components/schemas/AnthropicRateLimitError'
timeout_error:
$ref: '#/components/schemas/AnthropicGatewayTimeoutError'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicInvalidRequestError'
- $ref: '#/components/schemas/AnthropicAuthenticationError'
- $ref: '#/components/schemas/AnthropicBillingError'
- $ref: '#/components/schemas/AnthropicPermissionError'
- $ref: '#/components/schemas/AnthropicNotFoundError'
- $ref: '#/components/schemas/AnthropicRateLimitError'
- $ref: '#/components/schemas/AnthropicGatewayTimeoutError'
- $ref: '#/components/schemas/AnthropicAPIError'
- $ref: '#/components/schemas/AnthropicOverloadedError'
title: Error
request_id:
anyOf:
- type: string
- type: 'null'
default: null
title: Request Id
type:
const: error
default: error
title: Type
type: string
required:
- error
- request_id
- type
title: ErrorResponse
type: object
AnthropicModel:
type: string
title: Model
description: >-
The model that will complete your prompt. See the [Fireworks Model
Library](https://app.fireworks.ai/models) for available models.
AnthropicInputMessage:
additionalProperties: false
properties:
content:
anyOf:
- type: string
x-stainless-skip:
- go
- cli
- items:
$ref: '#/components/schemas/AnthropicInputContentBlock'
type: array
example:
- type: text
text: What is a quaternion?
title: Content
role:
enum:
- user
- assistant
title: Role
type: string
required:
- content
- role
title: InputMessage
type: object
discriminator:
propertyName: role
AnthropicMetadata:
additionalProperties: false
properties:
user_id:
anyOf:
- maxLength: 256
type: string
- type: 'null'
description: >-
An external identifier for the user who is associated with the
request.
This should be a uuid, hash value, or other opaque identifier. This
id may be used to help detect abuse. Do not include any identifying
information such as name, email address, or phone number.
examples:
- 13803d75-b4b5-4c3e-b2a2-6f21399b021b
title: User Id
title: Metadata
type: object
AnthropicOutputConfig:
additionalProperties: false
properties:
effort:
anyOf:
- $ref: '#/components/schemas/AnthropicEffortLevel'
- type: 'null'
description: >-
How much effort the model should put into its response. Higher
effort levels may result in more thorough analysis but take longer.
Valid values are `low`, `medium`, `high`, or `max`.
**Fireworks behavior:** The `effort` level is converted to Fireworks
[`reasoning_effort`](/api-reference/post-chatcompletions). Values
`low`, `medium`, and `high` map directly. `max` is mapped to `high`
since Fireworks does not have a `max` effort level.
format:
anyOf:
- $ref: '#/components/schemas/AnthropicJsonOutputFormat'
- type: 'null'
description: >-
A schema to specify the model's output format in responses. See
[structured
outputs](/structured-responses/structured-output-grammar-based)
title: OutputConfig
type: object
AnthropicRequestTextBlock:
additionalProperties: false
properties:
cache_control:
anyOf:
- discriminator:
mapping:
ephemeral:
$ref: '#/components/schemas/AnthropicCacheControlEphemeral'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicCacheControlEphemeral'
- type: 'null'
description: Create a cache control breakpoint at this content block.
title: Cache Control
citations:
anyOf:
- items:
discriminator:
mapping:
char_location:
$ref: >-
#/components/schemas/AnthropicRequestCharLocationCitation
content_block_location:
$ref: >-
#/components/schemas/AnthropicRequestContentBlockLocationCitation
page_location:
$ref: >-
#/components/schemas/AnthropicRequestPageLocationCitation
search_result_location:
$ref: >-
#/components/schemas/AnthropicRequestSearchResultLocationCitation
web_search_result_location:
$ref: >-
#/components/schemas/AnthropicRequestWebSearchResultLocationCitation
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicRequestCharLocationCitation'
- $ref: '#/components/schemas/AnthropicRequestPageLocationCitation'
- $ref: >-
#/components/schemas/AnthropicRequestContentBlockLocationCitation
- $ref: >-
#/components/schemas/AnthropicRequestWebSearchResultLocationCitation
- $ref: >-
#/components/schemas/AnthropicRequestSearchResultLocationCitation
type: array
- type: 'null'
title: Citations
text:
minLength: 1
title: Text
type: string
type:
const: text
title: Type
type: string
required:
- text
- type
title: RequestTextBlock
type: object
AnthropicThinkingConfigParam:
description: >-
Configuration for enabling the model's extended thinking.
When enabled, responses include `thinking` content blocks showing the
model's thinking process before the final answer. Requires a minimum
budget of 1,024 tokens and counts towards your `max_tokens` limit.
See [reasoning](/guides/reasoning) for details.
**Note:** The `adaptive` thinking type is not supported yet.
discriminator:
mapping:
adaptive:
$ref: '#/components/schemas/AnthropicThinkingConfigAdaptive'
disabled:
$ref: '#/components/schemas/AnthropicThinkingConfigDisabled'
enabled:
$ref: '#/components/schemas/AnthropicThinkingConfigEnabled'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicThinkingConfigEnabled'
- $ref: '#/components/schemas/AnthropicThinkingConfigDisabled'
- $ref: '#/components/schemas/AnthropicThinkingConfigAdaptive'
title: Thinking
AnthropicToolChoice:
description: >-
How the model should use the provided tools. The model can use a
specific tool, any available tool, decide by itself, or not use tools at
all.
discriminator:
mapping:
any:
$ref: '#/components/schemas/AnthropicToolChoiceAny'
auto:
$ref: '#/components/schemas/AnthropicToolChoiceAuto'
none:
$ref: '#/components/schemas/AnthropicToolChoiceNone'
tool:
$ref: '#/components/schemas/AnthropicToolChoiceTool'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicToolChoiceAuto'
- $ref: '#/components/schemas/AnthropicToolChoiceAny'
- $ref: '#/components/schemas/AnthropicToolChoiceTool'
- $ref: '#/components/schemas/AnthropicToolChoiceNone'
title: Tool Choice
AnthropicTool:
additionalProperties: false
properties:
type:
anyOf:
- type: 'null'
- const: custom
type: string
title: Type
description:
description: >-
Description of what this tool does.
Tool descriptions should be as detailed as possible. The more
information that the model has about what the tool is and how to use
it, the better it will perform. You can use natural language
descriptions to reinforce important aspects of the tool input JSON
schema.
examples:
- Get the current weather in a given location
title: Description
type: string
name:
description: >-
Name of the tool.
This is how the tool will be called by the model and in `tool_use`
blocks.
maxLength: 128
minLength: 1
pattern: ^[a-zA-Z0-9_-]{1,128}$
title: Name
type: string
input_schema:
$ref: '#/components/schemas/AnthropicInputSchema'
description: >-
[JSON schema](https://json-schema.org/draft/2020-12) for this tool's
input.
This defines the shape of the `input` that your tool accepts and
that the model will produce.
examples:
- properties:
location:
description: The city and state, e.g. San Francisco, CA
type: string
unit:
description: Unit for the output - one of (celsius, fahrenheit)
type: string
required:
- location
type: object
x-stainless-skip:
- cli
strict:
description: When true, guarantees schema validation on tool names and inputs
title: Strict
type: boolean
required:
- name
- input_schema
title: Tool
type: object
AnthropicContentBlock:
discriminator:
mapping:
redacted_thinking:
$ref: '#/components/schemas/AnthropicResponseRedactedThinkingBlock'
text:
$ref: '#/components/schemas/AnthropicResponseTextBlock'
thinking:
$ref: '#/components/schemas/AnthropicResponseThinkingBlock'
tool_use:
$ref: '#/components/schemas/AnthropicResponseToolUseBlock'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicResponseTextBlock'
- $ref: '#/components/schemas/AnthropicResponseThinkingBlock'
- $ref: '#/components/schemas/AnthropicResponseRedactedThinkingBlock'
- $ref: '#/components/schemas/AnthropicResponseToolUseBlock'
AnthropicStopReason:
enum:
- end_turn
- max_tokens
- stop_sequence
- tool_use
- pause_turn
- refusal
type: string
AnthropicRawOutput:
additionalProperties: false
description: >-
Fireworks extension that returns low-level details of what the model
sees, including the formatted prompt and function calls.
properties:
prompt_fragments:
description: >-
Pieces of the prompt (like individual messages) before truncation
and concatenation. Depending on prompt_truncate_len some of the
messages might be dropped. Contains a mix of strings to be tokenized
and individual tokens (if dictated by the conversation template)
items:
anyOf:
- type: string
- type: integer
title: Prompt Fragments
type: array
prompt_token_ids:
description: Fully processed prompt as seen by the model
items:
type: integer
title: Prompt Token Ids
type: array
completion:
description: >-
Raw completion produced by the model before any tool calls are
parsed
title: Completion
type: string
completion_token_ids:
anyOf:
- items:
type: integer
type: array
- type: 'null'
default: null
description: Token IDs for the raw completion
title: Completion Token Ids
images:
anyOf:
- items:
type: string
type: array
- type: 'null'
default: null
description: Images in the prompt
title: Images
grammar:
anyOf:
- type: string
- type: 'null'
default: null
description: >-
Grammar used for constrained decoding, can be either user provided
(directly or JSON schema) or inferred by the chat template
title: Grammar
required:
- prompt_fragments
- prompt_token_ids
- completion
title: RawOutput
type: object
AnthropicAPIError:
properties:
message:
default: Internal server error
title: Message
type: string
type:
const: api_error
default: api_error
title: Type
type: string
required:
- message
- type
title: APIError
type: object
AnthropicAuthenticationError:
properties:
message:
default: Authentication error
title: Message
type: string
type:
const: authentication_error
default: authentication_error
title: Type
type: string
required:
- message
- type
title: AuthenticationError
type: object
AnthropicBillingError:
properties:
message:
default: Billing error
title: Message
type: string
type:
const: billing_error
default: billing_error
title: Type
type: string
required:
- message
- type
title: BillingError
type: object
AnthropicInvalidRequestError:
properties:
message:
default: Invalid request
title: Message
type: string
type:
const: invalid_request_error
default: invalid_request_error
title: Type
type: string
required:
- message
- type
title: InvalidRequestError
type: object
AnthropicNotFoundError:
properties:
message:
default: Not found
title: Message
type: string
type:
const: not_found_error
default: not_found_error
title: Type
type: string
required:
- message
- type
title: NotFoundError
type: object
AnthropicOverloadedError:
properties:
message:
default: Overloaded
title: Message
type: string
type:
const: overloaded_error
default: overloaded_error
title: Type
type: string
required:
- message
- type
title: OverloadedError
type: object
AnthropicPermissionError:
properties:
message:
default: Permission denied
title: Message
type: string
type:
const: permission_error
default: permission_error
title: Type
type: string
required:
- message
- type
title: PermissionError
type: object
AnthropicRateLimitError:
properties:
message:
default: Rate limited
title: Message
type: string
type:
const: rate_limit_error
default: rate_limit_error
title: Type
type: string
required:
- message
- type
title: RateLimitError
type: object
AnthropicGatewayTimeoutError:
properties:
message:
default: Request timeout
title: Message
type: string
type:
const: timeout_error
default: timeout_error
title: Type
type: string
required:
- message
- type
title: GatewayTimeoutError
type: object
AnthropicInputContentBlock:
discriminator:
mapping:
document:
$ref: '#/components/schemas/AnthropicRequestDocumentBlock'
image:
$ref: '#/components/schemas/AnthropicRequestImageBlock'
redacted_thinking:
$ref: '#/components/schemas/AnthropicRequestRedactedThinkingBlock'
text:
$ref: '#/components/schemas/AnthropicRequestTextBlock'
thinking:
$ref: '#/components/schemas/AnthropicRequestThinkingBlock'
tool_result:
$ref: '#/components/schemas/AnthropicRequestToolResultBlock'
tool_use:
$ref: '#/components/schemas/AnthropicRequestToolUseBlock'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicRequestTextBlock'
description: Regular text content.
- $ref: '#/components/schemas/AnthropicRequestImageBlock'
description: >-
Image content specified directly as base64 data or as a reference
via a URL.
- $ref: '#/components/schemas/AnthropicRequestDocumentBlock'
description: >-
Document content, either specified directly as base64 data, as text,
or as a reference via a URL.
- $ref: '#/components/schemas/AnthropicRequestThinkingBlock'
description: A block specifying internal thinking by the model.
- $ref: '#/components/schemas/AnthropicRequestRedactedThinkingBlock'
description: A block specifying internal, redacted thinking by the model.
- $ref: '#/components/schemas/AnthropicRequestToolUseBlock'
description: A block indicating a tool use by the model.
- $ref: '#/components/schemas/AnthropicRequestToolResultBlock'
description: A block specifying the results of a tool use by the model.
x-stainless-python-extend-union:
- ContentBlock
x-stainless-python-extend-union-imports:
- from .content_block import ContentBlock
x-stainless-go-variant-constructor:
naming: new_{variant}_block
AnthropicEffortLevel:
description: All possible effort levels.
enum:
- low
- medium
- high
- max
title: EffortLevel
type: string
AnthropicJsonOutputFormat:
additionalProperties: false
properties:
schema:
additionalProperties: true
description: The JSON schema of the format
title: Schema
type: object
type:
const: json_schema
title: Type
type: string
required:
- schema
- type
title: JsonOutputFormat
type: object
AnthropicCacheControlEphemeral:
additionalProperties: false
properties:
ttl:
description: |-
The time-to-live for the cache control breakpoint.
This may be one the following values:
- `5m`: 5 minutes
- `1h`: 1 hour
Defaults to `5m`.
enum:
- 5m
- 1h
title: Ttl
type: string
x-stainless-renameMap:
ttl_5m: 5m
ttl_1h: 1h
type:
const: ephemeral
title: Type
type: string
required:
- type
title: CacheControlEphemeral
type: object
x-stainless-go-constant-constructor: true
AnthropicRequestCharLocationCitation:
additionalProperties: false
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- maxLength: 255
minLength: 1
type: string
- type: 'null'
title: Document Title
end_char_index:
title: End Char Index
type: integer
start_char_index:
minimum: 0
title: Start Char Index
type: integer
type:
const: char_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_char_index
- start_char_index
- type
title: RequestCharLocationCitation
type: object
AnthropicRequestContentBlockLocationCitation:
additionalProperties: false
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- maxLength: 255
minLength: 1
type: string
- type: 'null'
title: Document Title
end_block_index:
title: End Block Index
type: integer
start_block_index:
minimum: 0
title: Start Block Index
type: integer
type:
const: content_block_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_block_index
- start_block_index
- type
title: RequestContentBlockLocationCitation
type: object
AnthropicRequestPageLocationCitation:
additionalProperties: false
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- maxLength: 255
minLength: 1
type: string
- type: 'null'
title: Document Title
end_page_number:
title: End Page Number
type: integer
start_page_number:
minimum: 1
title: Start Page Number
type: integer
type:
const: page_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_page_number
- start_page_number
- type
title: RequestPageLocationCitation
type: object
AnthropicRequestSearchResultLocationCitation:
additionalProperties: false
properties:
cited_text:
title: Cited Text
type: string
end_block_index:
title: End Block Index
type: integer
search_result_index:
minimum: 0
title: Search Result Index
type: integer
source:
title: Source
type: string
start_block_index:
minimum: 0
title: Start Block Index
type: integer
title:
anyOf:
- type: string
- type: 'null'
title: Title
type:
const: search_result_location
title: Type
type: string
required:
- cited_text
- end_block_index
- search_result_index
- source
- start_block_index
- title
- type
title: RequestSearchResultLocationCitation
type: object
AnthropicRequestWebSearchResultLocationCitation:
additionalProperties: false
properties:
cited_text:
title: Cited Text
type: string
encrypted_index:
title: Encrypted Index
type: string
title:
anyOf:
- maxLength: 512
minLength: 1
type: string
- type: 'null'
title: Title
type:
const: web_search_result_location
title: Type
type: string
url:
maxLength: 2048
minLength: 1
title: Url
type: string
required:
- cited_text
- encrypted_index
- title
- type
- url
title: RequestWebSearchResultLocationCitation
type: object
AnthropicThinkingConfigAdaptive:
additionalProperties: false
properties:
type:
const: adaptive
title: Type
type: string
required:
- type
title: ThinkingConfigAdaptive
type: object
description: '**Not supported yet.**'
AnthropicThinkingConfigDisabled:
additionalProperties: false
properties:
type:
const: disabled
title: Type
type: string
required:
- type
title: ThinkingConfigDisabled
type: object
x-stainless-go-constant-constructor: true
AnthropicThinkingConfigEnabled:
additionalProperties: false
properties:
budget_tokens:
description: >-
Determines how many tokens the model can use for its internal
reasoning process. Larger budgets can enable more thorough analysis
for complex problems, improving response quality.
Must be ≥1024 and less than `max_tokens`.
See [reasoning](/guides/reasoning) for details.
minimum: 1024
title: Budget Tokens
type: integer
type:
const: enabled
title: Type
type: string
required:
- budget_tokens
- type
title: ThinkingConfigEnabled
type: object
AnthropicToolChoiceAny:
additionalProperties: false
description: The model will use any available tools.
properties:
disable_parallel_tool_use:
description: >-
Whether to disable parallel tool use.
Defaults to `false`. If set to `true`, the model will output exactly
one tool use.
title: Disable Parallel Tool Use
type: boolean
type:
const: any
title: Type
type: string
required:
- type
title: ToolChoiceAny
type: object
AnthropicToolChoiceAuto:
additionalProperties: false
description: The model will automatically decide whether to use tools.
properties:
disable_parallel_tool_use:
description: >-
Whether to disable parallel tool use.
Defaults to `false`. If set to `true`, the model will output at most
one tool use.
title: Disable Parallel Tool Use
type: boolean
type:
const: auto
title: Type
type: string
required:
- type
title: ToolChoiceAuto
type: object
AnthropicToolChoiceNone:
additionalProperties: false
description: The model will not be allowed to use tools.
properties:
type:
const: none
title: Type
type: string
required:
- type
title: ToolChoiceNone
type: object
x-stainless-go-constant-constructor: true
AnthropicToolChoiceTool:
additionalProperties: false
description: The model will use the specified tool with `tool_choice.name`.
properties:
disable_parallel_tool_use:
description: >-
Whether to disable parallel tool use.
Defaults to `false`. If set to `true`, the model will output exactly
one tool use.
title: Disable Parallel Tool Use
type: boolean
name:
description: The name of the tool to use.
title: Name
type: string
type:
const: tool
title: Type
type: string
required:
- name
- type
title: ToolChoiceTool
type: object
AnthropicInputSchema:
additionalProperties: true
properties:
properties:
anyOf:
- additionalProperties: true
type: object
- type: 'null'
title: Properties
required:
anyOf:
- items:
type: string
type: array
- type: 'null'
title: Required
type:
const: object
title: Type
type: string
required:
- type
title: InputSchema
type: object
AnthropicResponseRedactedThinkingBlock:
properties:
data:
title: Data
type: string
type:
const: redacted_thinking
default: redacted_thinking
title: Type
type: string
required:
- data
- type
title: ResponseRedactedThinkingBlock
type: object
AnthropicResponseTextBlock:
properties:
citations:
anyOf:
- items:
discriminator:
mapping:
char_location:
$ref: >-
#/components/schemas/AnthropicResponseCharLocationCitation
content_block_location:
$ref: >-
#/components/schemas/AnthropicResponseContentBlockLocationCitation
page_location:
$ref: >-
#/components/schemas/AnthropicResponsePageLocationCitation
search_result_location:
$ref: >-
#/components/schemas/AnthropicResponseSearchResultLocationCitation
web_search_result_location:
$ref: >-
#/components/schemas/AnthropicResponseWebSearchResultLocationCitation
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicResponseCharLocationCitation'
- $ref: '#/components/schemas/AnthropicResponsePageLocationCitation'
- $ref: >-
#/components/schemas/AnthropicResponseContentBlockLocationCitation
- $ref: >-
#/components/schemas/AnthropicResponseWebSearchResultLocationCitation
- $ref: >-
#/components/schemas/AnthropicResponseSearchResultLocationCitation
type: array
- type: 'null'
default: null
description: >-
Citations supporting the text block.
The type of citation returned will depend on the type of document
being cited. Citing a PDF results in `page_location`, plain text
results in `char_location`, and content document results in
`content_block_location`.
title: Citations
text:
maxLength: 5000000
minLength: 0
title: Text
type: string
type:
const: text
default: text
title: Type
type: string
required:
- citations
- text
- type
title: ResponseTextBlock
type: object
AnthropicResponseThinkingBlock:
properties:
signature:
title: Signature
type: string
thinking:
title: Thinking
type: string
type:
const: thinking
default: thinking
title: Type
type: string
required:
- signature
- thinking
- type
title: ResponseThinkingBlock
type: object
AnthropicResponseToolUseBlock:
properties:
id:
pattern: ^[a-zA-Z0-9_-]+$
title: Id
type: string
input:
additionalProperties: true
title: Input
type: object
name:
minLength: 1
title: Name
type: string
type:
const: tool_use
default: tool_use
title: Type
type: string
required:
- id
- input
- name
- type
title: ResponseToolUseBlock
type: object
AnthropicRequestDocumentBlock:
additionalProperties: false
properties:
cache_control:
anyOf:
- discriminator:
mapping:
ephemeral:
$ref: '#/components/schemas/AnthropicCacheControlEphemeral'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicCacheControlEphemeral'
- type: 'null'
description: Create a cache control breakpoint at this content block.
title: Cache Control
citations:
anyOf:
- $ref: '#/components/schemas/AnthropicRequestCitationsConfig'
- type: 'null'
context:
anyOf:
- minLength: 1
type: string
- type: 'null'
title: Context
source:
discriminator:
mapping:
base64:
$ref: '#/components/schemas/AnthropicBase64PDFSource'
content:
$ref: '#/components/schemas/AnthropicContentBlockSource'
text:
$ref: '#/components/schemas/AnthropicPlainTextSource'
url:
$ref: '#/components/schemas/AnthropicURLPDFSource'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicBase64PDFSource'
- $ref: '#/components/schemas/AnthropicPlainTextSource'
- $ref: '#/components/schemas/AnthropicContentBlockSource'
- $ref: '#/components/schemas/AnthropicURLPDFSource'
title: Source
title:
anyOf:
- maxLength: 500
minLength: 1
type: string
- type: 'null'
title: Title
type:
const: document
title: Type
type: string
required:
- source
- type
title: RequestDocumentBlock
type: object
AnthropicRequestImageBlock:
additionalProperties: false
properties:
cache_control:
anyOf:
- discriminator:
mapping:
ephemeral:
$ref: '#/components/schemas/AnthropicCacheControlEphemeral'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicCacheControlEphemeral'
- type: 'null'
description: Create a cache control breakpoint at this content block.
title: Cache Control
source:
discriminator:
mapping:
base64:
$ref: '#/components/schemas/AnthropicBase64ImageSource'
url:
$ref: '#/components/schemas/AnthropicURLImageSource'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicBase64ImageSource'
- $ref: '#/components/schemas/AnthropicURLImageSource'
title: Source
type:
const: image
title: Type
type: string
required:
- source
- type
title: RequestImageBlock
type: object
AnthropicRequestRedactedThinkingBlock:
additionalProperties: false
properties:
data:
title: Data
type: string
type:
const: redacted_thinking
title: Type
type: string
required:
- data
- type
title: RequestRedactedThinkingBlock
type: object
AnthropicRequestThinkingBlock:
additionalProperties: false
properties:
signature:
title: Signature
type: string
thinking:
title: Thinking
type: string
type:
const: thinking
title: Type
type: string
required:
- signature
- thinking
- type
title: RequestThinkingBlock
type: object
AnthropicRequestToolResultBlock:
additionalProperties: false
properties:
cache_control:
anyOf:
- discriminator:
mapping:
ephemeral:
$ref: '#/components/schemas/AnthropicCacheControlEphemeral'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicCacheControlEphemeral'
- type: 'null'
description: Create a cache control breakpoint at this content block.
title: Cache Control
content:
anyOf:
- type: string
x-stainless-skip:
- go
- cli
- items:
discriminator:
mapping:
document:
$ref: '#/components/schemas/AnthropicRequestDocumentBlock'
image:
$ref: '#/components/schemas/AnthropicRequestImageBlock'
text:
$ref: '#/components/schemas/AnthropicRequestTextBlock'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicRequestTextBlock'
- $ref: '#/components/schemas/AnthropicRequestImageBlock'
- $ref: '#/components/schemas/AnthropicRequestDocumentBlock'
title: Block
type: array
x-stainless-naming:
python:
type_name: Content
ruby:
type_name: Content
php:
type_name: Content
title: Content
is_error:
title: Is Error
type: boolean
tool_use_id:
pattern: ^[a-zA-Z0-9_-]+$
title: Tool Use Id
type: string
type:
const: tool_result
title: Type
type: string
required:
- tool_use_id
- type
title: RequestToolResultBlock
type: object
AnthropicRequestToolUseBlock:
additionalProperties: false
properties:
cache_control:
anyOf:
- discriminator:
mapping:
ephemeral:
$ref: '#/components/schemas/AnthropicCacheControlEphemeral'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicCacheControlEphemeral'
- type: 'null'
description: Create a cache control breakpoint at this content block.
title: Cache Control
id:
pattern: ^[a-zA-Z0-9_-]+$
title: Id
type: string
input:
additionalProperties: true
title: Input
type: object
name:
maxLength: 200
minLength: 1
title: Name
type: string
type:
const: tool_use
title: Type
type: string
required:
- id
- input
- name
- type
title: RequestToolUseBlock
type: object
AnthropicResponseCharLocationCitation:
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- type: string
- type: 'null'
title: Document Title
end_char_index:
title: End Char Index
type: integer
file_id:
anyOf:
- type: string
- type: 'null'
default: null
title: File Id
start_char_index:
minimum: 0
title: Start Char Index
type: integer
type:
const: char_location
default: char_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_char_index
- file_id
- start_char_index
- type
title: ResponseCharLocationCitation
type: object
AnthropicResponseContentBlockLocationCitation:
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- type: string
- type: 'null'
title: Document Title
end_block_index:
title: End Block Index
type: integer
file_id:
anyOf:
- type: string
- type: 'null'
default: null
title: File Id
start_block_index:
minimum: 0
title: Start Block Index
type: integer
type:
const: content_block_location
default: content_block_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_block_index
- file_id
- start_block_index
- type
title: ResponseContentBlockLocationCitation
type: object
AnthropicResponsePageLocationCitation:
properties:
cited_text:
title: Cited Text
type: string
document_index:
minimum: 0
title: Document Index
type: integer
document_title:
anyOf:
- type: string
- type: 'null'
title: Document Title
end_page_number:
title: End Page Number
type: integer
file_id:
anyOf:
- type: string
- type: 'null'
default: null
title: File Id
start_page_number:
minimum: 1
title: Start Page Number
type: integer
type:
const: page_location
default: page_location
title: Type
type: string
required:
- cited_text
- document_index
- document_title
- end_page_number
- file_id
- start_page_number
- type
title: ResponsePageLocationCitation
type: object
AnthropicResponseSearchResultLocationCitation:
properties:
cited_text:
title: Cited Text
type: string
end_block_index:
title: End Block Index
type: integer
search_result_index:
minimum: 0
title: Search Result Index
type: integer
source:
title: Source
type: string
start_block_index:
minimum: 0
title: Start Block Index
type: integer
title:
anyOf:
- type: string
- type: 'null'
title: Title
type:
const: search_result_location
default: search_result_location
title: Type
type: string
required:
- cited_text
- end_block_index
- search_result_index
- source
- start_block_index
- title
- type
title: ResponseSearchResultLocationCitation
type: object
AnthropicResponseWebSearchResultLocationCitation:
properties:
cited_text:
title: Cited Text
type: string
encrypted_index:
title: Encrypted Index
type: string
title:
anyOf:
- maxLength: 512
type: string
- type: 'null'
title: Title
type:
const: web_search_result_location
default: web_search_result_location
title: Type
type: string
url:
title: Url
type: string
required:
- cited_text
- encrypted_index
- title
- type
- url
title: ResponseWebSearchResultLocationCitation
type: object
AnthropicRequestCitationsConfig:
additionalProperties: false
properties:
enabled:
title: Enabled
type: boolean
title: RequestCitationsConfig
type: object
AnthropicBase64PDFSource:
additionalProperties: false
properties:
data:
format: byte
title: Data
type: string
media_type:
const: application/pdf
title: Media Type
type: string
type:
const: base64
title: Type
type: string
required:
- data
- media_type
- type
title: Base64PDFSource
type: object
AnthropicContentBlockSource:
additionalProperties: false
properties:
content:
anyOf:
- type: string
- items:
discriminator:
mapping:
image:
$ref: '#/components/schemas/AnthropicRequestImageBlock'
text:
$ref: '#/components/schemas/AnthropicRequestTextBlock'
propertyName: type
oneOf:
- $ref: '#/components/schemas/AnthropicRequestTextBlock'
- $ref: '#/components/schemas/AnthropicRequestImageBlock'
x-stainless-naming:
go:
type_name: ContentBlockSourceContentItem
title: content_block_source_content_item
type: array
title: content_block_source_content
title: Content
type:
const: content
title: Type
type: string
required:
- content
- type
title: ContentBlockSource
type: object
AnthropicPlainTextSource:
additionalProperties: false
properties:
data:
title: Data
type: string
media_type:
const: text/plain
title: Media Type
type: string
type:
const: text
title: Type
type: string
required:
- data
- media_type
- type
title: PlainTextSource
type: object
AnthropicURLPDFSource:
additionalProperties: false
properties:
type:
const: url
title: Type
type: string
url:
title: Url
type: string
required:
- type
- url
title: URLPDFSource
type: object
AnthropicBase64ImageSource:
additionalProperties: false
properties:
data:
format: byte
title: Data
type: string
media_type:
enum:
- image/jpeg
- image/png
- image/gif
- image/webp
title: Media Type
type: string
type:
const: base64
title: Type
type: string
required:
- data
- media_type
- type
title: Base64ImageSource
type: object
AnthropicURLImageSource:
additionalProperties: false
properties:
type:
const: url
title: Type
type: string
url:
title: Url
type: string
required:
- type
- url
title: URLImageSource
type: object
securitySchemes:
BearerAuth:
type: http
scheme: bearer
description: >-
Bearer authentication using your Fireworks API key. Format: Bearer
bearerFormat: API_KEY
````