Meta · First MLX build 18 Oct 2024
Llama-3.2-11B-Vision-Instruct
MLX
TextVision llama3.2
A 11B parameter vision and text model from Meta, with 5 MLX builds, from 4-bit to 8-bit.
- Parameters
- 11B From Hugging Face
- Quantisation
- 8-bit Llama-3.2-11B-Vision-Instruct-8bit
- Memory needed
- 15 GB Estimate, unified memory
- Download
- 11.3 GB Weights, estimated
- Downloads
- 2.8K 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 15 GB at 8-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-3.2-11B-Vision-Instruct-8bit --image photo.jpg --prompt "Describe this image"MLX builds
5 on mlx-community| Llama-3.2-11B-Vision-Instruct-4bit | 4-bit | 6.0 GB | 9 GB | 2.6K |
|---|---|---|---|---|
| Llama-3.2-11B-Vision-Instruct-abliterated-4-bit | 4-bit | 6.0 GB | 9 GB | 446 |
| Llama-3.2-11B-Vision-Instruct-8bit | 8-bit | 11.3 GB | 15 GB | 2.8K |
| Llama-3.2-11B-Vision-Instruct-abliterated-8-bit | 8-bit | 11.3 GB | 15 GB | 316 |
| Llama-3.2-11B-Vision-Instruct-abliterated | Unstated | 21.3 GB | 26 GB | 2.8K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.