Inference

    Endpoints

    Choose a published endpoint and connect an OpenAI-compatible client.

    Choose an endpoint

    Browse the published endpoint catalog by provider, tag, or use case. Popular production models can be called immediately; supported catalog models can also be deployed as dedicated endpoints.

    Open the endpoint catalog ↗Browse models →

    Copy the client values

    1. API Base URL

      Use the tenant-scoped URL displayed by the Console.

    2. Model Name

      Copy the exact published model identifier.

    3. API Key

      Use the key in the bearer authorization header.

    Validate the integration

    Run the sample REST call on the endpoint detail page or use the OpenAI-compatible client example before wiring the endpoint into your application.

    curl
    curl -X POST "$INFERX_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $INFERX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "YOUR_MODEL_ID",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'