OpenRelay API Documentation

From an API key to your first chat completion — unary and streaming — using nothing but curl.

This guide takes you from zero to your first completion in a couple of minutes.

Get an API key

Create an OpenRelay API key in the dashboard under Settings → API Keys. Keys are prefixed with vl_ and are shown in full only once — store it somewhere safe.

export OPENRELAY_API_KEY="vl_your_api_key"

See Authentication for the details.

Make your first request

Send a chat completion to https://inference.openrelay.inc/v1/chat/completions:

curl https://inference.openrelay.inc/v1/chat/completions \
  -H "Authorization: Bearer $OPENRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openrelay/gpt-oss-120b",
    "messages": [
      { "role": "user", "content": "Write a haiku about distributed systems." }
    ]
  }'

{
  "id": "chatcmpl-…",
  "object": "chat.completion",
  "model": "openrelay/gpt-oss-120b",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Nodes whisper in sync…" },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 16,
    "completion_tokens": 19,
    "total_tokens": 35
  }
}

The model field in the response is always the public id you requested.

Stream the response

Set "stream": true to receive the completion as Server-Sent Events. The stream ends with a final usage chunk and a data: [DONE] sentinel:

curl -N https://inference.openrelay.inc/v1/chat/completions \
  -H "Authorization: Bearer $OPENRELAY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openrelay/gpt-oss-120b",
    "stream": true,
    "messages": [
      { "role": "user", "content": "Count to five." }
    ]
  }'

data: {"id":"chatcmpl-…","object":"chat.completion.chunk","model":"openrelay/gpt-oss-120b","choices":[{"index":0,"delta":{"content":"1"}}]}

data: {"id":"chatcmpl-…","object":"chat.completion.chunk","model":"openrelay/gpt-oss-120b","choices":[{"index":0,"delta":{"content":" 2"}}]}

…

data: {"id":"chatcmpl-…","object":"chat.completion.chunk","model":"openrelay/gpt-oss-120b","choices":[],"usage":{"prompt_tokens":12,"completion_tokens":9,"total_tokens":21}}

data: [DONE]

-N disables curl's output buffering so you see tokens arrive live. See Streaming for the format details.

Next steps

Use your existing OpenAI client → Using OpenAI SDKs
Every request parameter and response field → Chat Completions
What's available and what it costs → Models and Pricing & billing

Quickstart

Get an API key

Make your first request

Stream the response

Next steps

On this page