4-bit Open on Hugging Face
Parameters
33B
Weights (est.)
≈ 18.4 GB
Mac memory (est.)
24 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/QwQ-32B-Preview-4bit --prompt "Hello"
Details
- Base model
- Qwen/QwQ-32B-Preview (quantized)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 59 in the last 30 days · 1.6K all time · 3 likes
- Published
- 2024-11-27 · updated 2024-11-27 (1y ago)
- Languages
- en
- Tags
- qwen2, chat
Other MLX builds of Qwen/QwQ-32B-Preview
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| QwQ-32B-Preview-2bit | 2-bit | ≈ 10.2 GB | 16 GB+ | 60 |
| QwQ-32B-Preview-3bit | 3-bit | ≈ 14.3 GB | 24 GB+ | 64 |
| QwQ-32B-Preview-4bit (this one) | 4-bit | ≈ 18.4 GB | 24 GB+ | 59 |
| Qwen_QwQ-32B-Preview_MLX-4bit | 4-bit | ≈ 18.4 GB | 24 GB+ | 56 |
| QwQ-32B-Preview-6bit | 6-bit | ≈ 26.6 GB | 32 GB+ | 81 |
| Qwen_QwQ-32B-Preview_MLX-8bit | 8-bit | ≈ 34.8 GB | 48 GB+ | 73 |
| QwQ-32B-Preview-8bit | 8-bit | ≈ 34.8 GB | 48 GB+ | 57 |
| QwQ-32B-Preview-bf16 | bf16 | ≈ 65.5 GB | 96 GB+ | 55 |