Skip to content

Aya Vision 32B

A multilingual vision language model with open weights

Visit official site

What Aya Vision 32B is

The model is described as a multilingual vision language model that takes image input and produces text output.

What you can do with it

  • Generate text from an image input
  • Try the model in a Hugging Face Space
  • Chat with the model through WhatsApp

Getting started

  • Install the needed transformers library
  • Load the model and the processor
  • Prepare image and text input data

Pros & cons

✓ What we like

  • Open weights are provided for researchers
  • OCR, image description and visual reasoning are listed

! What to watch out for

  • The model follows a CC-BY-NC license
  • Use is subject to the Cohere For AI usage policy

FAQ

How many languages are supported?

The card states support for 23 languages.

What is the context length?

The card lists a 16K context length.

Last reviewed: 2026-11-15

More AI image generation tools

View all →

How we review