MLX Community · First MLX build 30 Mar 2026
NVIDIA-Nemotron-3-Super-120B-A12B
MLX
Text nvidia-nemotron-open-model-license
A 121B parameter text generation model, with 4 MLX builds, from 4-bit to 8-bit.
- Parameters
- 121B From Hugging Face
- Quantisation
- 4-bit NVIDIA-Nemotron-3-Super-120B-A12B-4bit
- Memory needed
- 78 GB Estimate, unified memory
- Download
- 68.0 GB Weights, estimated
- Downloads
- 515 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 78 GB at 4-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/NVIDIA-Nemotron-3-Super-120B-A12B-4bit --prompt "Hello"MLX builds
4 on mlx-community| NVIDIA-Nemotron-3-Super-120B-A12B-4bit | 4-bit | 68.0 GB | 78 GB | 515 |
|---|---|---|---|---|
| NVIDIA-Nemotron-3-Super-120B-A12B-5bit | 5-bit | 83.1 GB | 95 GB | 233 |
| NVIDIA-Nemotron-3-Super-120B-A12B-6bit | 6-bit | 98.1 GB | 112 GB | 182 |
| NVIDIA-Nemotron-3-Super-120B-A12B-8bit | 8-bit | 128 GB | 146 GB | 146 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.