Microsoft · First MLX build 31 May 2026
Lens
MLX
Image mit
A 4.1B parameter text to image model from Microsoft, with 3 MLX builds, from 4-bit to bf16.
- Parameters
- 4.1B From Hugging Face
- Quantisation
- bf16 Lens-3.8B-bf16
- Memory needed
- 12 GB Estimate, unified memory
- Download
- 8.2 GB Weights, estimated
- Downloads
- 25 This build, last 30 days
- Task
- Text to image Image
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 12 GB at bf16
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/Lens-3.8B-bf16MLX builds
3 on mlx-community| Lens-3.8B-4bit | 4-bit | 2.3 GB | 5 GB | 13 |
|---|---|---|---|---|
| Lens-3.8B-8bit | 8-bit | 4.4 GB | 7 GB | 15 |
| Lens-3.8B-bf16 | bf16 | 8.2 GB | 12 GB | 25 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.