megvii-research · First MLX build 4 Jun 2026
NAFNet
MLX
Image mit
A 116M parameter image to image model from megvii-research, with 3 MLX builds.
- Parameters
- 116M From Hugging Face
- Quantisation
- Unstated NAFNet-GoPro-width64
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 272 MB Weights, estimated
- Downloads
- 69 This build, last 30 days
- Task
- Image to image Image
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at Unstated
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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/NAFNet-GoPro-width64MLX builds
3 on mlx-community| NAFNet-GoPro-width64 | Unstated | 272 MB | 3 GB | 69 |
|---|---|---|---|---|
| NAFNet-REDS-width64 | Unstated | 272 MB | 3 GB | 18 |
| NAFNet-SIDD-width64 | Unstated | 464 MB | 3 GB | 39 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.