On this page

Actions are the core functionality to interact with AI models.

## Inference

The inference action allows you to generate responses from AI models.

### Parameters

- `model` (string, required): Name of the model to use
- `context`(object, required): The context object containing:
  - `messages` (array): Array of message objects
  - `temperature` (number, optional): Sampling temperature, defaults to 1.0
  - `max_tokens` (number, optional): Maximum tokens to generate

### Returns

- `response`(object):
  - `text` (string): Generated response text
  - `usage` (object): Token usage statistics
  - `model` (string): Model used for inference

### Example

```text
POST /api/inference
Headers:
X-Vuedoo-Domain: your-domain
X-Vuedoo-Access-Key: your-access-key
{
  "model": "gpt-3.5-turbo",
  "context": {
    "messages": [\
      {"role": "user", "content": "Hello"}\
    ],
    "temperature": 0.7
  }
}
```