Alibaba Qwen · First MLX build 9 Jan 2025
Qwen2.5-0.5B-Instruct
MLX
Text apache-2.0
A 494M parameter text generation model from Alibaba Qwen, with 2 MLX builds at 2-bit.
- Parameters
- 494M From Hugging Face
- Quantisation
- 2-bit Qwen2.5-0.5B-Instruct-2bit-mlx
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 155 MB Weights, estimated
- Downloads
- 77 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 3 GB at 2-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-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen2.5-0.5B-Instruct-2bit-mlx --prompt "Hello"MLX builds
2 on mlx-community| Qwen2.5-0.5B-Instruct-2bit-mlx | 2-bit | 155 MB | 3 GB | 77 |
|---|---|---|---|---|
| qwen2.5-0.5b-instruct-q2 | 2-bit | 155 MB | 3 GB | 45 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.