One API, every Chinese model

OpenAI-compatible. Streaming supported. Most text, vision and agent models live here.

curl https://chinesemodels.ai/v1/chat/completions \
  -H "Authorization: Bearer $CHINESEMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-flash",
    "messages": [{ "role": "user", "content": "Hello" }],
    "stream": false
  }'

For clients built against the Claude SDK. Same models, Anthropic request body.

curl https://chinesemodels.ai/v1/messages \
  -H "Authorization: Bearer $CHINESEMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "GLM-5",
    "max_tokens": 256,
    "messages": [{ "role": "user", "content": "Hello" }]
  }'

Billed per generated image. Request fields follow the provider's own image API.

curl https://chinesemodels.ai/v1/images/generations \
  -H "Authorization: Bearer $CHINESEMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-3.0",
    "prompt": "a red panda on a mossy rock, morning light",
    "n": 1
  }'

Billed per task. Long-running: allow up to 15 minutes; the response carries a task id.

curl https://chinesemodels.ai/v1/videos \
  -H "Authorization: Bearer $CHINESEMODELS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Seedance-2.5",
    "prompt": "a corgi running through shallow water, 5s, 16:9",
    "duration": 5
  }'