Request#

The anthropic-version header is accepted and forwarded unchanged.

curl https://api.atptoken.ai/v1/messages \
  -H "Authorization: Bearer atp-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "<model from GET /v1/models>",
    "max_tokens": 256,
    "messages": [{"role": "user", "content": "hi"}]
  }'

Parameters#

The top-level system field accepts a string or an array of text blocks (cache_control on a block is accepted).

Field	Type	Description
model	string · required	A model id from GET /v1/models.
max_tokens	integer · required	Max output tokens (Anthropic requires this). With extended thinking, the thinking budget counts against it — too low returns a `200` with empty content (see Errors).
messages	array · required	Conversation turns, each { role, content }.
system	string or array	System prompt, as a string or an array of text blocks.
temperature	number	Sampling temperature, 0–1.
stream	boolean	Stream as Anthropic SSE events.

Next steps

Previous/v1/chat/completions Next/v1/models/{model}:generateContent

Was this page helpful?