DEVELOPER GUIDE

Make your first request

GPU Router speaks a strict, text-only subset of the OpenAI Chat Completions API.

Create an API key

Use your logged-in account for DB-earned credits. Wallet-authorized keys are only for the separately configured legacy rehearsal.

API Keys

Choose an exact model ID

Only models in the fresh gateway catalog can be routed.

Browse

Call the compatible endpoint

Use your existing OpenAI client with a different base URL.

YOU'RE HERE

STAGED

The integration contract is ready, but live inference admission is currently disabled on this network.

OPENAI COMPATIBLE

Chat Completions

Model

These examples pin the selected billing source. Earned-only requests do not charge prepaid funds. Deferred batches with a billing assertion are not supported yet.

curl 'https://gpurouter.io/v1/chat/completions' \
  -H "Authorization: Bearer $GPU_ROUTER_API_KEY" \
  -H 'Content-Type: application/json' \
  -H 'x-gpu-billing: {"version":1,"source":"prepaid"}' \
  --data-binary '{
  "model": "your-model-id",
  "messages": [
    {
      "role": "user",
      "content": "Hello!"
    }
  ],
  "max_tokens": 256
}'

Routing

The gateway estimates request cost and chooses the cheapest eligible primary route for this exact model ID.

Metering

It reserves worst-case credits before dispatch, then settles or refunds against reported usage.

Streaming

Set stream: true; the gateway requests an upstream usage frame for final settlement.