Messages
Updated: Nov 11, 2026
POST
See API Guide - Authentication.
See API Guide - Common Request Headers.
| Header | Required | Description |
|---|---|---|
anthropic-version | No | Anthropic API version. The official Anthropic SDK sends it automatically. |
application/json- body
model*stringModel ID, e.g. deepseek-v4-flash.
messages*array<>List of conversation messages; each item has role and content.
max_tokens*integerMaximum number of tokens to generate.
Range
1 <= valuesystem?|array<>System prompt, passed separately from messages.
temperature?numberSampling temperature.
Range
0 <= value <= 1top_p?numberNucleus sampling parameter.
top_k?integerTop-K sampling parameter.
stream?booleanWhether to stream the response.
stop_sequences?array<string>Stop sequences.
tools?array<>Tool definitions.
tool_choice?Tool choice strategy.
thinking?Thinking / reasoning configuration.
metadata?Metadata.
Successful response
application/json- response
id?stringMessage ID.
type?stringObject type.
role?stringMessage role.
content?array<>List of content blocks.
model?stringID of the model actually used.
stop_reason?stringStop reason.
usage?Token usage statistics.
curl -X POST 'https://api.cysuo.com/v1/messages' \ -H 'Content-Type: application/json' \ -d '{ "model": "deepseek-v4-flash", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello, please introduce yourself." } ]}'Example Response200
{ "id": "string", "type": "string", "role": "string", "content": [ { "type": "string", "text": "string" } ], "model": "string", "stop_reason": "string", "usage": { "input_tokens": 0, "output_tokens": 0, "cache_creation_input_tokens": 0, "cache_read_input_tokens": 0 }}