Skip to main content
If you already use the OpenAI SDK, the only change is pointing base_url at OrcaRouter and using an OrcaRouter API key.

​
1. Get an API key

Sign in at orcarouter.ai and copy your key from the dashboard. OrcaRouter keys start with sk-orca-. See Get an API key for the options you can set when creating a key (credit limit, expiration).

​
2. Make a call

curl https://api.orcarouter.ai/v1/chat/completions \
  -H "Authorization: Bearer sk-orca-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o-mini",
    "messages": [{"role":"user","content":"Hello, OrcaRouter!"}]
  }'
That’s it. Everything else — streaming, tools, vision, JSON mode, Claude models, Gemini models, Grok models — works exactly as it would against the provider directly. Swap openai/gpt-4o-mini for any model ID you see in /v1/models. Models are namespaced by provider: openai/gpt-4o-mini, anthropic/claude-sonnet-4.6, google/gemini-2.5-pro, deepseek/deepseek-chat, grok/grok-4-fast-reasoning, qwen/qwen3.6-plus, kimi/kimi-k2.6, minimax/minimax-m2.7, etc.

​
Next steps

Web search

Let the model search the web before answering.

Auto Router

Let OrcaRouter pick the cheapest live model.

Model Fallbacks

Chain models for resilience.

Chat Completions

Full API reference with try-it.