aiDex API

Call any AI model (or your whole team) from a single endpoint.

curl

curl -X POST https://aidex.chat/api/v1/chat/completions \
  -H "Authorization: Bearer aidx_sk_your_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4.6",
    "messages": [{"role": "user", "content": "Hello from the API!"}],
    "stream": true
  }'

OpenAI-compatible

Drop-in replacement for /v1/chat/completions. Existing client libraries work out of the box.

5 chat modes

Solo, Team, Compare, Judge, Pipeline, every mode the web app offers, available via REST + SSE.

Same billing

API calls use the same BYOK keys and managed-access system. No hidden costs.

Built-in rate limiting

Per-key RPM limits with X-RateLimit headers. Scales with your plan.

Real-time streaming

SSE streaming with OpenAI-compatible delta chunks. Latency-sensitive apps work out of the box.

Web search included

Pass web_search: true and the model browses the web, same native-search routing as the UI.

Rate limits by plan

Plan	RPM	Max keys
BYOK	20	3
BYOK + Add-on	40	5
Pro	60	10
Power	120	25