AI API GATEWAY
Access the Best AI Models for Free
A clean AI API console for plans, gateway status, and usage analytics.
# Fully compatible with the OpenAI SDK $ curl https://gateway.tokenav.ai/v1/chat/completions \ -H "Authorization: Bearer $TOKENAV_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "gpt-5", "messages": [{ "role": "user", "content": "Hello!" }] }' # Response { "id": "chatcmpl_9f2c", "usage": { "total_tokens": 217 } }
FEATURES
Why Developers Choose Tokenav
The simplest way to integrate AI into your application
Per-request metering
Every call records its model, tokens and cost, labeled with the transport it used. You always know where the money goes.
Sample data
Context pool
Long-session context lives on the gateway, so repeated prefixes are never billed twice. The longer the conversation, the more you save.
WebSocket upstream
WS keep-alive mode cuts handshake overhead. Even plain HTTP clients get bridged to WS on the gateway side.
Multi-upstream failover
When one upstream fails, traffic switches to a healthy line automatically. Your calls keep going.
One-line integration
Compatible with the OpenAI API format. No refactor needed, just swap the base_url.
Three steps to integrate.
01
Create a key
Sign up and generate an API key in the console. Top up anytime, pay as you go.
02
Change one line
Point the request URL at Tokenav. Everything else stays exactly the same.
03
Start calling
Use your favorite SDK, Codex, Claude Code or any tool. Usage records update in real time.
Grab your key first.
Free credits · No credit card required