Base URL for OpenAI Compatable Things:

Introduction

Sylph provides an OpenAI-compatible API for interacting with various AI models across different providers. Our API follows RESTful practices and returns JSON responses.

GET /health

Check API health status and provider availability.

Request 
curl
Response 
{
  "status": "ok",
  "providers": {
    "available": ["HackClubProvider", "GoogleProvider", "..."],
    "disabled": [],
    "errors": {}
  }
}

Authentication

Authentication is optional but recommended. You can provide any string as the API key:

HTTP Header 
Authorization: Bearer any-string-works

Models

Browse all available models at /models

GET /v1/models

List all available models across enabled providers.

Request 
curl /v1/models \
     -H "Authorization: Bearer any-string"
Response 
{
  "object": "list",
  "data": [
    {
      "id": "hackclub/llama-3.3-70b-versatile",
      "object": "model",
      "created": 1743648062945,
      "owned_by": "https://hackclub.com",
      "permission": [],
      "root": "llama",
      "context_length": 128000,
      "capabilities": {
        "text": true,
        "images": false
      },
      "health": {
        "status": "operational",
        "latency": 312
      }
    }
  ]
}

Chat

POST /v1/chat/completions

Create a chat completion with the specified model.

Request Parameters

Parameter Type Description
model* string ID of the model to use
messages* array Array of messages for the conversation
stream boolean Whether to stream responses (default: false)
temperature number Controls randomness (0-2, default: 0.7)
Request 
curl /v1/chat/completions \
     -H "Content-Type: application/json" \
     -H "Authorization: Bearer any-string" \
     -d '{
  "model": "hackclub/llama-3.3-70b-versatile",
  "messages": [
    {"role": "user", "content": "Hello!"}
  ],
  "temperature": 0.7
}'
Response 
{
  "id": "chat-12345",
  "object": "chat.completion",
  "created": 1743648062945,
  "model": "hackclub/llama-3.3-70b-versatile",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 10,
    "completion_tokens": 8,
    "total_tokens": 18
  }
}

Errors

Code Description
400 Invalid request parameters
401 Invalid or missing API key
404 Model not found
429 Too many requests
500 Server or provider error
Error Response Example 
{
  "error": {
    "message": "Model not found: invalid-model",
    "type": "invalid_request_error",
    "code": "model_not_found",
    "param": "model"
  }
}