GET

/v1/schemas/baseten/chat/v1/messages

Accept:

kind inputactivity chathash c939077a…extracted Oct 6, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `aa7bde20942ca711969ec6b81aab7d14ac913e23ef78ca5424d48038046d9da7` { "fetchedAt": 1791264068, "extractorVersion": "1", "verifiedAt": null }

Request body

12 properties · 3 required · 6 $defs

property	type	description	constraints
model*	string	The model slug to use. Find available models at [Model APIs](https://app.baseten.co/model-apis/create).
messages*	InputMessage[]	The conversation history as an ordered list of input messages. Alternating `user` and `assistant` roles are expected; the final message must be from the `user`.
max_tokens*	integer	The maximum number of tokens to generate in the response. Required by the Messages API. The response may be shorter if it finishes naturally or hits a stop sequence.	n ≥ 1
system	string \| TextBlock[]	A system prompt that sets the model's behavior. Pass either a single string or an array of text content blocks.
temperature	number	Controls randomness. Lower values are more deterministic. Range: 0 to 1.	0 ≤ n ≤ 1 · default: 1
top_p	number	Nucleus sampling: only consider tokens with cumulative probability up to this value.	0 < n ≤ 1
top_k	integer	Limits token selection to the top K most probable tokens at each step.	n ≥ 0
stop_sequences	string[]	Custom text sequences that will stop generation. When a stop sequence is hit, `stop_reason` is `stop_sequence` and `stop_sequence` contains the matched string.
stream	boolean	If `true`, the response is streamed as server-sent events. Each event has a `type` such as `message_start`, `content_block_delta`, or `message_stop`.	default: false
tools	ToolDefinition[]	A list of tools the model may call. Each tool has a `name`, `description`, and `input_schema` (a JSON Schema object).
tool_choice	object \| object \| object \| …	Controls which tool (if any) the model must call.
metadata	object	An object describing metadata about the request. Supports `user_id` for abuse detection.

Types

6 $defs

InputMessageobject

A message in the conversation.

role*	enum	The role of the message author. System prompts are passed via the top-level `system` field, not as a message.	user \| assistant
content*	string \| TextBlock \| ToolUseBlock \| ToolResultBlock[]	The message content. Pass a string for text-only messages or an array of content blocks for messages that include tool calls or tool results.

TextBlockobject

A text content block.

type*	const	The block type, always `text`.	"text"
text*	string	The text content.

ToolChoiceobject | object | object | …

Controls which tool (if any) the model calls.

ToolDefinitionobject

A tool the model can call.

name*	string	A unique name for the tool.
description	string	A description of what the tool does. The model uses this to decide when to call it.
input_schema*	object	A JSON Schema object describing the tool's expected input.

ToolResultBlockobject

A tool result content block supplied by the user in response to a tool call.

type*	const	The block type, always `tool_result`.	"tool_result"
tool_use_id*	string	The `id` of the `tool_use` block this result responds to.
content*	string \| TextBlock[]	The tool's output. Pass a string for simple results or an array of text blocks for structured output.
is_error	boolean	Set to `true` if the tool call raised an error.	default: false

ToolUseBlockobject

A tool use content block generated by the model.

type*	const	The block type, always `tool_use`.	"tool_use"
id*	string	A unique identifier for this tool call. Use this as `tool_use_id` in the matching tool result.
name*	string	The name of the tool being called.
input*	object	The arguments passed to the tool, matching the tool's `input_schema`.

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"baseten","endpointId":"v1/messages","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}