bf16 Open on Hugging Face
Parameters
31B
Weights (est.)
≈ 61.1 GB
Mac memory (est.)
96 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/SU-01-bf16 --prompt "Hello"
Details
- Base model
- Simplified-Reasoning/SU-01 (finetune)
- Vendor
- Simplified-Reasoning
- Task
- Text generation
- Quantisation
- bf16
- Licence
- apache-2.0
- Downloads
- 15 in the last 30 days · 125 all time · 0 likes
- Published
- 2026-05-16 · updated 2026-05-16 (5mo ago)
- Tags
- qwen3_moe, reasoning, olympiad, mathematics, science, reinforcement-learning, test-time-scaling, long-context
Other MLX builds of Simplified-Reasoning/SU-01
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| SU-01-mxfp4 | MXFP4 | ≈ 17.2 GB | 24 GB+ | 19 |
| SU-01-4bit | 4-bit | ≈ 17.2 GB | 24 GB+ | 17 |
| SU-01-nvfp4 | 4-bit | ≈ 17.2 GB | 24 GB+ | 15 |
| SU-01-5bit | 5-bit | ≈ 21.0 GB | 32 GB+ | 17 |
| SU-01-6bit | 6-bit | ≈ 24.8 GB | 32 GB+ | 12 |
| SU-01-8bit | 8-bit | ≈ 32.4 GB | 48 GB+ | 18 |
| SU-01-mxfp8 | MXFP8 | ≈ 32.4 GB | 48 GB+ | 14 |
| SU-01-bf16 (this one) | bf16 | ≈ 61.1 GB | 96 GB+ | 15 |