DeepSeek-R1-Distill-Qwen-14B-3bit
mlx-community/DeepSeek-R1-Distill-Qwen-14B-3bit
Uncategorised · mlx
3-bit Open on Hugging Face
Parameters
15B
Weights (est.)
≈ 6.5 GB
Mac memory (est.)
16 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-14B (quantized)
- Vendor
- DeepSeek
- Task
- Uncategorised
- Quantisation
- 3-bit
- Licence
- —
- Downloads
- 275 in the last 30 days · 52K all time · 7 likes
- Published
- 2025-01-20 · updated 2025-02-26 (1y ago)
- Tags
- qwen2
Other MLX builds of deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| DeepSeek-R1-Distill-Qwen-14B-3bit (this one) | 3-bit | ≈ 6.5 GB | 16 GB+ | 275 |
| DeepSeek-R1-Distill-Qwen-14B-4bit | 4-bit | ≈ 8.3 GB | 16 GB+ | 2.4K |
| DeepSeek-R1-Distill-Qwen-14B-6bit | 6-bit | ≈ 12.0 GB | 16 GB+ | 729 |
| DeepSeek-R1-Distill-Qwen-14B-8bit | 8-bit | ≈ 15.7 GB | 24 GB+ | 709 |
| DeepSeek-R1-Distill-Qwen-14B | — | ≈ 29.5 GB | 48 GB+ | 188 |
| DeepSeek-R1-Distill-Qwen-14B-bf16 | bf16 | ≈ 29.5 GB | 48 GB+ | 56 |
| DeepSeek-R1-Distill-Qwen-14B-MLX | — | — | — | 0 |