MLX Community · First MLX build 24 Dec 2024
QVQ-72B-Preview
MLX
TextVision qwen
A 73B parameter vision and text model, with 5 MLX builds, from 3-bit to bf16.
- Parameters
- 73B From Hugging Face
- Quantisation
- 4-bit QVQ-72B-Preview-4bit
- Memory needed
- 49 GB Estimate, unified memory
- Download
- 41.3 GB Weights, estimated
- Downloads
- 29 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 49 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/QVQ-72B-Preview-4bit --image photo.jpg --prompt "Describe this image"MLX builds
5 on mlx-community| QVQ-72B-Preview-3bit | 3-bit | 32.1 GB | 38 GB | 21 |
|---|---|---|---|---|
| QVQ-72B-Preview-4bit | 4-bit | 41.3 GB | 49 GB | 29 |
| QVQ-72B-Preview-6bit | 6-bit | 59.6 GB | 69 GB | 19 |
| QVQ-72B-Preview-8bit | 8-bit | 78.0 GB | 90 GB | 24 |
| QVQ-72B-Preview-bf16 | bf16 | 147 GB | 167 GB | 24 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.