> ## Documentation Index
> Fetch the complete documentation index at: https://praison.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Vision Inference • AI Agents Framework

> vision_inference: API reference for TrainVisionModel.vision_inference

# vision\_inference

<div className="flex items-center gap-2">
  <Badge color="purple">Method</Badge>
</div>

> This is a method of the [**TrainVisionModel**](../classes/TrainVisionModel) class in the [**train\_vision**](../modules/train_vision) module.

## Signature

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
def vision_inference(instruction: Any, image: Any) -> Any
```

## Parameters

<ParamField query="instruction" type="Any" required={true}>
  No description available.
</ParamField>

<ParamField query="image" type="Any" required={true}>
  No description available.
</ParamField>

### Returns

<ResponseField name="Returns" type="Any">
  The result of the operation.
</ResponseField>

## Uses

* `FastVisionModel.for_inference`
* `apply_chat_template`
* `to`
* `hf_tokenizer`
* `generate`
* `batch_decode`

## Used By

* [`main`](../functions/main)

## Source

<Card title="View on GitHub" icon="github" href="https://github.com/MervinPraison/PraisonAI/blob/main/src/praisonai/praisonai/train_vision.py#L194">
  `praisonai/train_vision.py` at line 194
</Card>

***

## Related Documentation

<CardGroup cols={2}>
  <Card title="Multimodal" icon="eye" href="/docs/features/multimodal" />
</CardGroup>