MLX Community · First MLX build 23 Jul 2025
Qwen3-235B-A22B-Instruct-2507
MLX
Text apache-2.0
A 235B parameter text generation model, with 2 MLX builds, from 3-bit DWQ to 4-bit DWQ.
- Parameters
- 235B From Hugging Face
- Quantisation
- 3-bit DWQ Qwen3-235B-A22B-Instruct-2507-3bit-DWQ
- Memory needed
- 117 GB Estimate, unified memory
- Download
- 103 GB Weights, estimated
- Downloads
- 94 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 117 GB at 3-bit DWQ
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/Qwen3-235B-A22B-Instruct-2507-3bit-DWQ --prompt "Hello"MLX builds
2 on mlx-community| Qwen3-235B-A22B-Instruct-2507-3bit-DWQ | 3-bit DWQ | 103 GB | 117 GB | 94 |
|---|---|---|---|---|
| Qwen3-235B-A22B-Instruct-2507-4bit-DWQ | 4-bit DWQ | 132 GB | 150 GB | 62 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.