Gateway API

The gateway is a transparent reverse proxy. Point your LLM client at Nolma’s gateway URL instead of the provider’s URL directly.

Base URL

https://gateway.nolma.ai

Authentication

Every request must include:

Header	Required	Description
`NM-Key`	Yes	Your Nolma API key (`nm_live_...`)
`NM-Agent`	Recommended	Agent name for tracking
`NM-Session`	Optional	Session ID (SDK sets automatically)
`NM-Env`	Optional	`production` / `staging` / `development`
`NM-User`	Optional	End-user ID for per-user budgets

Provider routing

Provider	Gateway URL
OpenAI	`https://gateway.nolma.ai/openai/v1/chat/completions`
Anthropic	`https://gateway.nolma.ai/anthropic/v1/messages`
Gemini	`https://gateway.nolma.ai/gemini/v1/models/{model}:generateContent`
Groq	`https://gateway.nolma.ai/groq/openai/v1/chat/completions`
Mistral	`https://gateway.nolma.ai/mistral/v1/chat/completions`

The gateway forwards the entire request body unchanged. The response format is identical to the provider’s native response.

Example

curl -X POST https://gateway.nolma.ai/anthropic/v1/messages \
  -H "NM-Key: nm_live_abc123" \
  -H "NM-Agent: my-agent" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-4-5-20251001",
    "max_tokens": 100,
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Health check

GET https://gateway.nolma.ai/health

Returns:

{"status": "ok", "version": "0.1.0"}

​Gateway API

​Base URL

​Authentication

​Provider routing

​Example

​Health check