Qwen2.5-7B-Instruct-1M-bf16
mlx-community/Qwen2.5-7B-Instruct-1M-bf16
Text generation · transformers
bf16 Open on Hugging Face
Parameters
7.6B
Weights (est.)
≈ 15.2 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/Qwen2.5-7B-Instruct-1M-bf16 --prompt "Hello"
Details
- Base model
- Qwen/Qwen2.5-7B-Instruct-1M (finetune)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- bf16
- Licence
- apache-2.0
- Downloads
- 75 in the last 30 days · 512 all time · 0 likes
- Published
- 2025-01-26 · updated 2025-01-26 (1y ago)
- Languages
- en
- Tags
- qwen2, chat
Other MLX builds of Qwen/Qwen2.5-7B-Instruct-1M
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen2.5-7B-Instruct-1M-3bit | 3-bit | ≈ 3.3 GB | 8 GB+ | 109 |
| Qwen2.5-7B-Instruct-1M-4bit | 4-bit | ≈ 4.3 GB | 8 GB+ | 254 |
| Qwen2.5-7B-Instruct-1M-6bit | 6-bit | ≈ 6.2 GB | 16 GB+ | 109 |
| Qwen2.5-7B-Instruct-1M-8bit | 8-bit | ≈ 8.1 GB | 16 GB+ | 124 |
| Qwen2.5-7B-Instruct-1M-bf16 (this one) | bf16 | ≈ 15.2 GB | 24 GB+ | 75 |