4-bit Open on Hugging Face
Parameters
235B
Weights (est.)
≈ 132 GB
Mac memory (est.)
192 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/Qwen3-235B-A22B-4bit --prompt "Hello"
Details
- Base model
- Qwen/Qwen3-235B-A22B (quantized)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 4.7K in the last 30 days · 315K all time · 9 likes
- Published
- 2025-04-29 · updated 2025-04-29 (1y ago)
- Tags
- qwen3_moe
Other MLX builds of Qwen/Qwen3-235B-A22B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen3-235B-A22B-3bit | 3-bit | ≈ 103 GB | 128 GB+ | 41 |
| Qwen3-235B-A22B-3bit-DWQ | 3-bit DWQ | ≈ 103 GB | 128 GB+ | 38 |
| Qwen3-235B-A22B-4bit (this one) | 4-bit | ≈ 132 GB | 192 GB+ | 4.7K |
| Qwen3-235B-A22B-mixed-3-4bit | 4-bit | ≈ 132 GB | 192 GB+ | 23 |
| Qwen3-235B-A22B-mixed-3-6bit | 6-bit | ≈ 191 GB | 256 GB+ | 24 |
| Qwen3-235B-A22B-8bit | 8-bit | ≈ 250 GB | 512 GB+ | 1.3K |