Alibaba Qwen · First MLX build 16 Feb 2026
Qwen3.5-397B-A17B
MLX
Text apache-2.0
A 397B parameter text generation model from Alibaba Qwen, with 2 MLX builds, from 4-bit to 8-bit.
- Parameters
- 397B From Hugging Face
- Quantisation
- 4-bit Qwen3.5-397B-A17B-nvfp4
- Memory needed
- 252 GB Estimate, unified memory
- Download
- 223 GB Weights, estimated
- Downloads
- 1.6K 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 252 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-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen3.5-397B-A17B-nvfp4 --prompt "Hello"MLX builds
2 on mlx-community| Qwen3.5-397B-A17B-nvfp4 | 4-bit | 223 GB | 252 GB | 1.6K |
|---|---|---|---|---|
| Qwen3.5-397B-A17B-8bit | 8-bit | 422 GB | 474 GB | 161 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.