OPENAI COMPATIBLE

API reference

Use an OpenAI SDK or make HTTPS requests directly. Your API key is the credential for every request.

Authentication

Bearer authentication is recommended. The API also accepts x-api-key and x-goog-api-key for compatible clients.

Authorization: Bearer $HANROUTER_API_KEYx-api-key: $HANROUTER_API_KEYx-goog-api-key: $HANROUTER_API_KEY

Manage API keys

Core endpoints

GET/v1/models

List the model IDs currently available to this key.

POST/v1/chat/completions

Generate a model response from a message list.

Direct HTTP request

Use the same Base URL and API key without an SDK.

GET /v1/models

curl https://www.hanrouter.com/v1/models \
  -H "Authorization: Bearer $HANROUTER_API_KEY"

POST /v1/chat/completions

curl https://www.hanrouter.com/v1/chat/completions \
  -H "Authorization: Bearer $HANROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-model-id",
    "messages": [
      { "role": "user", "content": "Explain this API in one sentence." }
    ],
    "temperature": 0.2
  }'

Streaming response

Set stream to true to receive Server-Sent Events. Read each delta as it arrives and wait for the final completion event.

cURL

curl -N https://www.hanrouter.com/v1/chat/completions \
  -H "Authorization: Bearer $HANROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-model-id",
    "messages": [{ "role": "user", "content": "Write a short greeting." }],
    "stream": true
  }'

Errors and retries

Do not retry authentication failures. For rate-limit or transient upstream failures, use exponential backoff and preserve the request ID in your logs.

401Check that the key is active, complete, and sent in one supported header.

429Reduce concurrency, wait briefly, then retry with exponential backoff.

5xxRetry only idempotent or safely repeatable work after a short delay.