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

# AI Transforms

> Leverage LLMs to transform your data

<Note>
  This feature is currently in private beta. Please reach out to us on Slack if
  you're interested in trying it out
</Note>

## Overview

AI Transforms can be used to augment data models using large language models in either the cloud or locally. An `LLMTransform` handles token management, column metadata context and more that is provided via Vinyl internals.

Lets take some user reviews of movies from the IMDB dataset. Suppose we want to tag each review with sentiment analysis using the latest LLM models available.

| id | movie\_title        | comment\_id | comment                                                           |
| -- | ------------------- | ----------- | ----------------------------------------------------------------- |
| 1  | A Quiet Place       | 101         | An intense experience with a few plot holes.                      |
| 2  | The Room            | 102         | So bad its almost good. A cult classic for all the wrong reasons. |
| 3  | Lost in Translation | 103         | Subtly powerful and emotionally complex.                          |
| 4  | Requiem for a Dream | 104         | Disturbing content that leaves a lasting impression.              |

We can import the `LLMTransform` and `OpenAIProvider` from Vinyl to run an LLM-powered transform on a specific column of our dataset.

```python models.py theme={null}
from my_project.sources.taxi_sample import TaxiSample
from vinyl.ai import LLMTransform, OpenAIProvider

@model(deps=Movies)
def movie_sentiment(m):
    infer_sentiment = LLMTransform(
        OpenAIProvider,
        cols=[
            m.comment,
        ]
        prompt="Rate the sentiment of this movie review comment column either 'positive', 'neutral' or 'negative'"
    )

    m.mutate({
        "sentiment": infer_sentiment()
    })

    return t
```

Vinyl handles concurrency, caching and incremental updates. Vinyl will only run the transform on new rows. The `OpenAIProvider` allows you to bring your own OpenAI API key.

We can preview the results of this using:

```bash theme={null}
vinyl preview model movie_sentiment
```

Here are 4 rows of revised example data with movie titles and comments that better align with their respective sentiment labels:

| id | movie\_title        | comment\_id | comment                                                           | sentiment |
| -- | ------------------- | ----------- | ----------------------------------------------------------------- | --------- |
| 1  | A Quiet Place       | 101         | An intense experience with a few plot holes.                      | Neutral   |
| 2  | The Room            | 102         | So bad its almost good. A cult classic for all the wrong reasons. | Negative  |
| 3  | Lost in Translation | 103         | Subtly powerful and emotionally complex.                          | Positive  |
| 4  | Requiem for a Dream | 104         | Disturbing content that leaves a lasting impression.              | Negative  |

## Local AI Inference

Working with data workflows comes with a host of privacy challenges. If you're interested in running AI Transforms using self-hosted models and GPU inference [reach out to us](https://www.turntable.so/join-waitlist?source=ai-transform).