GET

/v1/schemas/baseten/chat/v1/messages?kind=output

kind outputactivity chathash 9ca1d72b…extracted Oct 6, 2026version currentsource upstream spec ↗
fieldrecorded source
schemaSource linked above.

upstream-spec

Source trace
aa7bde20942ca711969ec6b81aab7d14ac913e23ef78ca5424d48038046d9da7
{
  "fetchedAt": 1791264068,
  "extractorVersion": "1",
  "verifiedAt": null
}

Response body

8 properties · 7 required · 3 $defs
propertytypedescriptionconstraints
id*stringA unique identifier for this message, such as `msg_abc123`.
type*constThe object type, always `message`."message"
role*constThe role of the generated message, always `assistant`."assistant"
content*TextBlock | ToolUseBlock[]An array of content blocks generated by the model. Text responses contain a single `text` block; responses that invoke tools contain `tool_use` blocks.
model*stringThe model slug that produced the response.
stop_reason*enumWhy the model stopped generating: `end_turn` (natural stop), `max_tokens` (hit the `max_tokens` limit), `stop_sequence` (matched a `stop_sequences` entry), or `tool_use` (model invoked a tool).end_turn | max_tokens | stop_sequence | tool_use
usage*UsageToken usage statistics for the request.
stop_sequencestringThe stop sequence that was matched, if `stop_reason` is `stop_sequence`. Otherwise `null`.

Types

3 $defs

TextBlockobject

A text content block.

type*constThe block type, always `text`."text"
text*stringThe text content.

ToolUseBlockobject

A tool use content block generated by the model.

type*constThe block type, always `tool_use`."tool_use"
id*stringA unique identifier for this tool call. Use this as `tool_use_id` in the matching tool result.
name*stringThe name of the tool being called.
input*objectThe arguments passed to the tool, matching the tool's `input_schema`.

Usageobject

Token usage statistics.

input_tokensintegerThe number of input tokens processed, excluding tokens served from cache. Total prompt size is `input_tokens + cache_read_input_tokens + cache_creation_input_tokens`.
output_tokensintegerThe number of output tokens generated.
cache_read_input_tokensintegerThe number of input tokens served from cache. Not counted in `input_tokens`, matching the Anthropic API: `input_tokens` covers only uncached tokens. Caching is automatic; no `cache_control` blocks are required. This field is omitted when no tokens were read fr…

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", …}]}