Chat / Text

Anthropic Protocol (Claude)

Anthropic Messages API compatible. For Claude models (GPT model names also work via auto protocol translation).

Endpoints

Usage	Method	Path
Chat (REST + SSE)	POST	/v1/messages
Token estimate (free)	POST	/v1/messages/count_tokens

Authentication

Authorization: Bearer <API_KEY>
Content-Type: application/json

Supported Models

claude-* (e.g. claude-haiku-4-5-20251001); gpt-* models are also supported on /v1/messages — the gateway auto-translates between Anthropic and OpenAI protocols. See GET /v1/models.

Request Parameters

Param	Type	Required	Description
model	string	Required	Model, e.g. claude-haiku-4-5-20251001
messages	array	Required	Messages (role/content)
max_tokens	integer	Required	Max output tokens
system	string/array	—	System prompt
stream	boolean	—	true for SSE streaming
temperature / top_p / top_k	number	—	Sampling controls
tools / tool_choice	—	—	Tool calling

Streaming (SSE)

Set "stream": true and receive SSE with curl -N. Events: message_start → content_block_delta → … → message_delta(usage) → message_stop.

Request Example

Non-streaming

curl -X POST "https://api.cqtai.com/v1/messages" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-4-5-20251001",
    "max_tokens": 1024,
    "messages": [{"role":"user","content":"你好，介绍一下你自己"}]
  }'

Streaming (SSE)

curl -N -X POST "https://api.cqtai.com/v1/messages" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-4-5-20251001",
    "max_tokens": 1024,
    "stream": true,
    "messages": [{"role":"user","content":"写一首关于夏天的短诗"}]
  }'

Token estimate (free)

curl -X POST "https://api.cqtai.com/v1/messages/count_tokens" \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-4-5-20251001",
    "messages": [{"role":"user","content":"估算这段话的 token"}]
  }'

Response Example

{
  "id": "msg_xxx",
  "type": "message",
  "role": "assistant",
  "model": "claude-haiku-4-5-20251001",
  "content": [{ "type": "text", "text": "你好！我是 Claude……" }],
  "stop_reason": "end_turn",
  "usage": { "input_tokens": 12, "output_tokens": 28 }
}

Billing & Credits

Billed per token: input × input price + output × output price (cache write ×1.25, cache read ×0.1), then × your rate. count_tokens is free. See the Intro page for model prices.