MLX Community · First MLX build 4 Jun 2026
Nemotron-3-Ultra-550B-A55B
MLX
Text nvidia-open-model-license Trending
A 549B parameter text generation model, with 3 MLX builds, from 4-bit to bf16.
- Parameters
- 549B From Hugging Face
- Quantisation
- 4-bit Nemotron-3-Ultra-550B-A55B-4bit
- Memory needed
- 352 GB Estimate, unified memory
- Download
- 313 GB Weights, estimated
- Downloads
- 339 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 352 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/Nemotron-3-Ultra-550B-A55B-4bit --prompt "Hello"MLX builds
3 on mlx-community| Nemotron-3-Ultra-550B-A55B-4bit | 4-bit | 313 GB | 352 GB | 339 |
|---|---|---|---|---|
| Nemotron-3-Ultra-550B-A55B | 4-bit | 313 GB | 352 GB | 253 |
| Nemotron-3-Ultra-550B-A55B-bf16 | bf16 | — | — | 0 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.