8-bit Open on Hugging Face
Parameters
15B
Weights (est.)
≈ 15.7 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/Qwen2.5-Coder-14B-8bit --prompt "Hello"
Details
- Base model
- Qwen/Qwen2.5-Coder-14B (quantized)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- 8-bit
- Licence
- apache-2.0
- Downloads
- 171 in the last 30 days · 1.2K all time · 0 likes
- Published
- 2024-11-11 · updated 2024-11-12 (2y ago)
- Languages
- en
- Tags
- qwen2, code, qwen, qwen-coder, codeqwen
Other MLX builds of Qwen/Qwen2.5-Coder-14B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen2.5-Coder-14B-4bit | 4-bit | ≈ 8.3 GB | 16 GB+ | 750 |
| Qwen2.5-Coder-14B-8bit (this one) | 8-bit | ≈ 15.7 GB | 24 GB+ | 171 |
| Qwen2.5-Coder-14B-bf16 | bf16 | ≈ 29.5 GB | 48 GB+ | 171 |