Custom OpenAI-Compatible Providers Are Now Supported

You can now register custom OpenAI-compatible providers in LLM Gateway. Perfect for internal deployments or specialized third-party APIs that speak the OpenAI Chat Completions format.

Configure a Custom Provider

Add a provider in the UI (lowercase name, base URL, and token). Then call models via {providerName}/{modelName}:

curl -X POST "https://api.llmgateway.io/v1/chat/completions" \
  -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mycompany/custom-gpt-4",
    "messages": [{"role": "user", "content": "Hello from my custom provider!"}]
  }'

curl -X POST "https://api.llmgateway.io/v1/chat/completions" \
  -H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mycompany/custom-gpt-4",
    "messages": [{"role": "user", "content": "Hello from my custom provider!"}]
  }'

Requirements include a lowercase provider name and a valid HTTPS base URL. See details in the docs: Custom Providers.