bf16 Open on Hugging Face
Parameters
8.2B
Weights (est.)
≈ 16.4 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/Qwen3-8B-Base-bf16 --prompt "Hello"
Details
- Base model
- Qwen/Qwen3-8B-Base (finetune)
- Vendor
- Qwen
- Task
- Text generation
- Quantisation
- bf16
- Licence
- apache-2.0
- Downloads
- 141 in the last 30 days · 748 all time · 1 likes
- Published
- 2025-05-03 · updated 2025-05-03 (1y ago)
- Tags
- qwen3
Other MLX builds of Qwen/Qwen3-8B-Base
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Qwen3-8B-Base-bf16 (this one) | bf16 | ≈ 16.4 GB | 24 GB+ | 141 |
| Qwen3-Embedding-8B-mxfp8 | MXFP8 | — | — | 1.0K |
| Qwen3-Reranker-8B-mxfp8 | MXFP8 | — | — | 231 |