GET

/v1/schemas/minimax/chat/anthropic/v1/messages?kind=output

Accept:

kind outputactivity chathash 73af9238…extracted Nov 8, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `c41e967945ea1159466901032e21618b0da87cb1063ee2f3c244cc6e998b8752` { "fetchedAt": 1794115838, "extractorVersion": "2", "verifiedAt": null }

Response body

7 properties · 0 required · 2 $defs

property	type	description	constraints
id	string	Unique ID of this response
type	enum	Object type, fixed as `message`	message
role	enum	Role, fixed as `assistant`	assistant
model	string	Model ID used for this request
content	ResponseContentBlock[]	List of response content blocks
stop_reason	enum	Reason for stopping generation: - end_turn: Model ended naturally - max_tokens: Reached max_tokens limit - tool_use: Model requested tool use - pause_turn: A server tool (such as web_search) reached its usage limit and the turn was paused. Append the returned …	end_turn \| max_tokens \| tool_use \| pause_turn
usage	Usage	Token usage for this request, including prompt cache usage when applicable.

Types

2 $defs

ResponseContentBlockobject

Content block returned in response messages. The model emits text, tool_use (when calling tools), and thinking (when extended thinking is active). Responses never contain image, video, tool_result, or mid_conv_system blocks.

type*	enum	Content block type: - text: Text content - tool_use: Model tool call - thinking: Model thinking content	text \| tool_use \| thinking
text	string	Text content when type=text.
id	string	Tool call ID returned when type=tool_use.
name	string	Tool name returned when type=tool_use.
input	object	Tool input returned when type=tool_use.
thinking	string	Model thinking content when type=thinking.
signature	string	Signature for thinking content. Return unchanged in later turns.

Usageobject

Token usage for this request, including prompt cache usage when applicable.

input_tokens	integer	Number of tokens consumed by input
output_tokens	integer	Number of tokens consumed by output
cache_creation_input_tokens	integer	Number of input tokens used to create prompt cache entries.
cache_read_input_tokens	integer	Number of input tokens read from prompt cache.

Validate a payload

before you spend tokens

shell

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