Hundreds of models
Frontier and open-weight models from every major lab, behind a single OpenAI-compatible endpoint. Swap a model string, not an integration.
chatstreamingtoolsvisionjson mode
Point your OpenAI SDK at air router and reach hundreds of models instantly. Pay as you go from a prepaid balance, with transparent per-model pricing.
curl https://api.airrouter.app/api/public/v1/chat/completions \
-H "Authorization: Bearer sk-gw-..." \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-4o-mini",
"messages": [{"role":"user","content":"Hello"}]
}'one endpoint · every major lab
what you get
Frontier and open-weight models from every major lab, behind a single OpenAI-compatible endpoint. Swap a model string, not an integration.
Top up with a card, spend down per token. No subscriptions, no invoices, no surprise bills.
Every request logged with tokens, latency, model and exact cost — filterable by API key.
Issue as many keys as you need, revoke instantly. Stored hashed, never in plaintext.
about a minute
Email or Google. No sales call, no waitlist, no credit card to look around.
Buy a prepaid pack with a card. Your balance drains per token, and only per token.
Point your existing OpenAI client at air router, keep every other line of code.
pricing
Every model lists its per-million-token price with our routing fee already included. What you see in the catalog is what leaves your balance.
See the full catalog →Create an account, add credits, generate a key, change one base URL.
Create an account