Alibaba Qwen · First MLX build 16 Feb 2025
Qwen2-VL-7B-Instruct
MLX
TextVision apache-2.0
A 8.3B parameter vision and text model from Alibaba Qwen, with 4 MLX builds, from 4-bit to 8-bit.
- Parameters
- 8.3B From Hugging Face
- Quantisation
- 8-bit Qwen2-VL-7B-Instruct-abliterated-8bit
- Memory needed
- 12 GB Estimate, unified memory
- Download
- 8.8 GB Weights, estimated
- Downloads
- 72 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 12 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/Qwen2-VL-7B-Instruct-abliterated-8bit --image photo.jpg --prompt "Describe this image"MLX builds
4 on mlx-community| Qwen2-VL-7B-Instruct-abliterated-4bit | 4-bit | 4.7 GB | 8 GB | 66 |
|---|---|---|---|---|
| Qwen2-VL-7B-Instruct-abliterated-6bit | 6-bit | 6.7 GB | 10 GB | 50 |
| Qwen2-VL-7B-Instruct-abliterated-8bit | 8-bit | 8.8 GB | 12 GB | 72 |
| Qwen2-VL-7B-Instruct-abliterated | Unstated | 16.6 GB | 21 GB | 51 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.