4-bit Open on Hugging Face
Parameters
7.6B
Weights (est.)
≈ 4.3 GB
Mac memory (est.)
8 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-7B-4bit --prompt "Hello"
Details
- Base model
- Qwen/Qwen2.5-7B (finetune)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 1.2K in the last 30 days · 5.1K all time · 0 likes
- Published
- 2024-09-18 · updated 2024-09-18 (2y ago)
- Languages
- en
- Tags
- qwen2, code, qwen, qwen-coder, codeqwen
Other MLX builds of Qwen/Qwen2.5-7B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen2.5-7B-Instruct-4bit | 4-bit | ≈ 4.3 GB | 8 GB+ | 13K |
| Qwen2.5-Coder-7B-4bit (this one) | 4-bit | ≈ 4.3 GB | 8 GB+ | 1.2K |
| Qwen2.5-Coder-7B-8bit | 8-bit | ≈ 8.1 GB | 16 GB+ | 1.0K |
| Qwen2.5-7B-Instruct-8bit | 8-bit | ≈ 8.1 GB | 16 GB+ | 293 |
| A.X-4.0-Light-MLX-BF16 | bf16 | ≈ 14.5 GB | 24 GB+ | 38 |
| Qwen2.5-Coder-7B-bf16 | bf16 | ≈ 15.2 GB | 24 GB+ | 206 |
| Qwen2.5-7B-Instruct-bf16 | bf16 | ≈ 15.2 GB | 24 GB+ | 166 |