Integration

Integrating your application with UsageGuard involves creating a connection in UsageGuard Dashboard or through Management API, once you have a connection, you can start sending requests to UsageGuard.

Unified Inference API

Inference API is our recommended API that allows you to leverage UsageGuard's built-in safeguards and moderation features, as well as access to multiple models without changing your existing code.

Here's an example of how to format a request to the Unified Inference API:

{
  "model": "claude-3.5-sonnet:latest",
  "messages": [
    {
      "role": "user",
      "content": "Translate the following English text to French: 'Hello, how are you?'"
    }
  ],
  "parameters": {
    "maxTokens": 60,
    "temperature": 0.7
  }
}

To use this API, you would send a POST request to the endpoint with your UsageGuard API key in the header:

POST /v1/inference/chat HTTP/1.1
Host: api.usageguard.com
x-ug-api-key: Your_UsageGuard_API_key
Content-Type: application/json

This structure allows you to specify the model, provide the conversation context in the messages array, and set various parameters to control the model's output.