MLX Community · First MLX build 5 Mar 2025
aya-vision-32b
MLX
TextVision cc-by-nc-4.0
A 33B parameter vision and text model, with 5 MLX builds, from 3-bit to bf16.
- Parameters
- 33B From Hugging Face
- Quantisation
- 3-bit aya-vision-32b-3bit
- Memory needed
- 19 GB Estimate, unified memory
- Download
- 15.1 GB Weights, estimated
- Downloads
- 44 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 19 GB at 3-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-vlm$ pip install -U mlx-vlm
$ python -m mlx_vlm.generate --model mlx-community/aya-vision-32b-3bit --image photo.jpg --prompt "Describe this image"MLX builds
5 on mlx-community| aya-vision-32b-3bit | 3-bit | 15.1 GB | 19 GB | 44 |
|---|---|---|---|---|
| aya-vision-32b-4bit | 4-bit | 19.2 GB | 24 GB | 32 |
| aya-vision-32b-6bit | 6-bit | 27.4 GB | 33 GB | 28 |
| aya-vision-32b-8bit | 8-bit | 35.6 GB | 42 GB | 25 |
| aya-vision-32b-bf16 | bf16 | 66.3 GB | 77 GB | 37 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.