MLX Community · First MLX build 16 Feb 2026
dots.ocr
MLX
TextVision mit
A 3B parameter vision and text model, with 8 MLX builds, from 4-bit to bf16.
- Parameters
- 3B From Hugging Face
- Quantisation
- 8-bit dots.ocr-8bit
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 4.4 GB Weights, estimated
- Downloads
- 181 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 8 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 7 GB at 8-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-vlm$ pip install -U mlx-vlm
$ python -m mlx_vlm.generate --model mlx-community/dots.ocr-8bit --image photo.jpg --prompt "Describe this image"MLX builds
8 on mlx-community| dots.ocr-4bit | 4-bit | 3.5 GB | 6 GB | 163 |
|---|---|---|---|---|
| dots.ocr-mxfp4 | MXFP4 | 3.5 GB | 6 GB | 36 |
| dots.ocr-nvfp4 | 4-bit | 3.5 GB | 6 GB | 34 |
| dots.ocr-5bit | 5-bit | 3.7 GB | 7 GB | 31 |
| dots.ocr-6bit | 6-bit | 4.0 GB | 7 GB | 25 |
| dots.ocr-8bit | 8-bit | 4.4 GB | 7 GB | 181 |
| dots.ocr-mxfp8 | MXFP8 | 4.4 GB | 7 GB | 38 |
| dots.ocr-bf16 | bf16 | 6.1 GB | 9 GB | 115 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.