Alibaba Qwen · First MLX build 14 Dec 2024
Qwen2.5-Coder-7B-Instruct
MLX
Text apache-2.0
A 7.6B parameter text generation model from Alibaba Qwen, with 2 MLX builds at 3-bit.
- Parameters
- 7.6B From Hugging Face
- Quantisation
- Unstated Qwen2.5-Coder-7B-Instruct-MLX
- Memory needed
- 20 GB Estimate, unified memory
- Download
- 15.2 GB Weights, estimated
- Downloads
- 1.3K This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 20 GB at Unstated
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-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen2.5-Coder-7B-Instruct-MLX --prompt "Hello"MLX builds
2 on mlx-community| Qwen2.5-Coder-7B-Instruct-3bit | 3-bit | 3.3 GB | 6 GB | 326 |
|---|---|---|---|---|
| Qwen2.5-Coder-7B-Instruct-MLX | Unstated | 15.2 GB | 20 GB | 1.3K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.