Downloads · 30 days
67
9% of all-time downloads
Nemotron-3.5-Lightning-30B-A3B-MLX-2bit
majentik/Nemotron-3.5-Lightning-30B-A3B-MLX-2bit
Nemotron-3.5-Lightning-30B-A3B-MLX-2bit is a text generation model from majentik. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as other.
MLX 2bit (affine, group size 32) quantized variant of nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 for Apple silicon via mlx-lm.
All-time downloads
725
Public
Parameters
31.6B
11.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors11.9 GB · 100%
Parameter types
How the weights are stored.
U3231.6B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Nemotron-3.5-Lightning-30B-A3B-MLX-2bit
MLX 2bit (affine, group size 32) quantized variant of nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 for Apple silicon via mlx-lm.
Provenance
- Source: nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 @ revision
d468880b6ad3c6e0d21377ce7242adaea4cc884d(OpenMDW v1.1 — see LICENSE in this repo). - Quantized with
mlx_lm.convert(mlx-lm 0.31.3): affine, 2-bit, group size 32.
Smoke gate
Before upload this pack passed a deterministic coherence gate: greedy
48-token chat generation loaded through
mlx_lm.load, judged for emptiness, repetition loops, multi-script
gibberish, and special-token debris. Verdict: ok.
Usage
pip install mlx-lm
mlx_lm.generate --model majentik/Nemotron-3.5-Lightning-30B-A3B-MLX-2bit --prompt "Hello"
Evaluation
| Benchmark | Score |
|---|---|
| arc_easy_acc | 0.4050 |
| hellaswag_acc | 0.3150 |
License
OpenMDW v1.1 (openmdw-1.1): permissive open model, distribution, and
weights license; the full license text ships in this repository as
LICENSE. See https://openmdw.ai/license/1-1/.
Available tiers
- Task
- Text Generation
- Library
- mlx
- Type
- nemotron_h
- License
- other
- Created
- Aug 15, 2026
- Updated
- Aug 17, 2026