QuickstartOpenAI CompatibleAPI Key
ChinaWHAPI API Quickstart Guide
Call DeepSeek, Qwen, Kimi, GLM, Doubao and other Chinese LLMs through a single OpenAI-compatible interface.
API Endpoint
ChinaWHAPI uses OpenAI-compatible request format, allowing developers to call multiple Chinese mainstream LLMs through a unified gateway. For production, it is recommended to store API Keys in server environment variables rather than in frontend code.
POST https://chinawhapi.com/v1/chat/completions
Minimal Request Example
Replace 'model' with the enabled model name in the console, and replace 'API Key' with your user key.
curl https://chinawhapi.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer {your_api_key}" \
-d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"Hello"}]}'Getting Your API Key
Log in to the ChinaWHAPI console and generate a new key on the API Keys page. Copy and securely store it; do not hardcode in frontend code.
- Console URL: https://chinawhapi.com/console
- Supports multiple key management for easy project isolation
- Can set key expiration time and permission scope
Common Checks
If a request fails, first verify the key, model name, account balance, and subscription status.
- 401 usually means invalid API Key
- 403 usually means insufficient balance or expired plan
- 400 usually means incorrect request body or model name
- 429 usually means rate limit triggered