Skip to content

Extracting entities from images

Marvin can use OpenAI's vision API to process images and convert them into structured data, transforming unstructured information into native types that are appropriate for a variety of programmatic use cases.

The marvin.beta.extract function is an enhanced version of marvin.extract that accepts images as well as text.


Please note that vision support in Marvin is still in beta, as OpenAI has not finalized the vision API yet. While it works as expected, it is subject to change.

What it does

The beta extract function can extract entities from images and text.

How it works

This involves a two-step process: first, a caption is generated for the image that is aligned with the structuring goal. Next, the actual extract operation is performed with an LLM.

Example: identifying dogs

We will extract the breed of each dog in this image:

import marvin

img = marvin.beta.Image(

result = marvin.beta.extract(img, target=str, instructions="dog breeds")


result == ["Pembroke Welsh Corgi", "Yorkshire Terrier"]

Model parameters

You can pass parameters to the underlying API via the model_kwargs and vision_model_kwargs arguments of extract. These parameters are passed directly to the respective APIs, so you can use any supported parameter.

Async support

If you are using Marvin in an async environment, you can use extract_async:

result = await marvin.beta.extract_async(
    "I drove from New York to California.",
    instructions="2-letter state codes",

assert result == ["NY", "CA"]