Meta · First MLX build 19 Dec 2024
Llama-3.2-90B-Vision-Instruct
MLX
TextVision llama3.2
A 89B parameter vision and text model from Meta, with one MLX build at 4-bit.
- Parameters
- 89B From Hugging Face
- Quantisation
- 4-bit Llama-3.2-90B-Vision-Instruct-4bit
- Memory needed
- 58 GB Estimate, unified memory
- Download
- 49.8 GB Weights, estimated
- Downloads
- 775 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 58 GB at 4-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/Llama-3.2-90B-Vision-Instruct-4bit --image photo.jpg --prompt "Describe this image"MLX builds
1 on mlx-community| Llama-3.2-90B-Vision-Instruct-4bit | 4-bit | 49.8 GB | 58 GB | 775 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.