DeepSeek-R1-Distill-Qwen-32B-3bit
mlx-community/DeepSeek-R1-Distill-Qwen-32B-3bit
Uncategorised · mlx
3-bit Open on Hugging Face
Parameters
33B
Weights (est.)
≈ 14.3 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.
Details
- Base model
- deepseek-ai/DeepSeek-R1-Distill-Qwen-32B (quantized)
- Vendor
- DeepSeek
- Task
- Uncategorised
- Quantisation
- 3-bit
- Licence
- —
- Downloads
- 134 in the last 30 days · 30K all time · 4 likes
- Published
- 2025-01-20 · updated 2025-02-26 (1y ago)
- Tags
- qwen2
Other MLX builds of deepseek-ai/DeepSeek-R1-Distill-Qwen-32B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-32B-3bit (this one) | 3-bit | ≈ 14.3 GB | 24 GB+ | 134 |
| DeepSeek-R1-Distill-Qwen-32B-4bit | 4-bit | ≈ 18.4 GB | 24 GB+ | 5.2K |
| DeepSeek-R1-Distill-Qwen-32B-MLX-4Bit | 4-bit | ≈ 18.4 GB | 24 GB+ | 724 |
| DeepSeek-R1-Distill-Qwen-32B-Q | 4-bit | ≈ 18.4 GB | 24 GB+ | 93 |
| DeepSeek-R1-Distill-Qwen-32B-6bit | 6-bit | ≈ 26.6 GB | 32 GB+ | 201 |
| DeepSeek-R1-Distill-Qwen-32B-MLX-8Bit | 8-bit | ≈ 34.8 GB | 48 GB+ | 1.5K |
| DeepSeek-R1-Distill-Qwen-32B | — | ≈ 65.5 GB | 96 GB+ | 234 |
| DeepSeek-R1-Distill-Qwen-32B-bf16 | bf16 | ≈ 65.5 GB | 96 GB+ | 113 |
| DeepSeek-R1-Distill-Qwen-32B-float16 | — | ≈ 65.5 GB | 96 GB+ | 64 |
| DeepSeek-R1-Distill-Qwen-32B-MLX | — | — | — | 0 |