8-bit Open on Hugging Face
Parameters
32B
Weights (est.)
≈ 33.6 GB
Mac memory (est.)
48 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/Nemotron-Cascade-2-30B-A3B-8bit --prompt "Hello"
Details
- Base model
- nvidia/Nemotron-Cascade-2-30B-A3B (quantized)
- Vendor
- NVIDIA
- Task
- Text generation
- Quantisation
- 8-bit
- Licence
- nvidia-open-model-license
- Downloads
- 252 in the last 30 days · 4.4K all time · 8 likes
- Published
- 2026-03-20 · updated 2026-03-20 (7mo ago)
- Languages
- en
- Tags
- nemotron_h, nvidia, nemotron-cascade-2, reasoning, general-purpose, SFT, RL, custom_code
Other MLX builds of nvidia/Nemotron-Cascade-2-30B-A3B
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| Nemotron-Cascade-2-30B-A3B-4bit | 4-bit | ≈ 17.8 GB | 24 GB+ | 628 |
| Nemotron-Cascade-2-30B-A3B-nvfp4 | 4-bit | ≈ 17.8 GB | 24 GB+ | 153 |
| Nemotron-Cascade-2-30B-A3B-mxfp4 | MXFP4 | ≈ 17.8 GB | 24 GB+ | 39 |
| Nemotron-Cascade-2-30B-A3B-5bit | 5-bit | ≈ 21.7 GB | 32 GB+ | 41 |
| Nemotron-Cascade-2-30B-A3B-6bit | 6-bit | ≈ 25.7 GB | 32 GB+ | 209 |
| Nemotron-Cascade-2-30B-A3B-mlx-6bit | 6-bit | ≈ 25.7 GB | 32 GB+ | 44 |
| Nemotron-Cascade-2-30B-A3B-8bit (this one) | 8-bit | ≈ 33.6 GB | 48 GB+ | 252 |
| Nemotron-Cascade-2-30B-A3B-mxfp8 | MXFP8 | ≈ 33.6 GB | 48 GB+ | 28 |
| Nemotron-Cascade-2-30B-A3B-bf16 | bf16 | ≈ 63.2 GB | 96 GB+ | 37 |