thinkingmachines · First MLX build 30 Jul 2026
Inkling-Small
MLX
TextVision apache-2.0
A 264B parameter vision and text model from thinkingmachines, with 4 MLX builds, from 2-bit to 4-bit.
- Parameters
- 264B From Hugging Face
- Quantisation
- 2-bit Inkling-Small-mlx-2bit
- Memory needed
- 101 GB Estimate, unified memory
- Download
- 88.4 GB Weights, estimated
- Downloads
- 295 This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 101 GB at 2-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/Inkling-Small-mlx-2bit --image photo.jpg --prompt "Describe this image"MLX builds
4 on mlx-community| Inkling-Small-mlx-2bit | 2-bit | 88.4 GB | 101 GB | 295 |
|---|---|---|---|---|
| Inkling-Small-mlx-3bit | 3-bit | 121 GB | 138 GB | 118 |
| Inkling-Small-mxfp4 | MXFP4 | 149 GB | 168 GB | 196 |
| Inkling-Small-mlx-4bit | 4-bit | 153 GB | 174 GB | 248 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.