IBM · First MLX build 2 Oct 2025
granite-4.0-h-tiny
MLX
Text apache-2.0
A 6.9B parameter text generation model from IBM, with 5 MLX builds, from 3-bit to 8-bit.
- Parameters
- 6.9B From Hugging Face
- Quantisation
- 4-bit DWQ Granite-4.0-H-Tiny-4bit-DWQ
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 3.9 GB Weights, estimated
- Downloads
- 144 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 7 GB at 4-bit DWQ
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/Granite-4.0-H-Tiny-4bit-DWQ --prompt "Hello"MLX builds
5 on mlx-community| granite-4.0-h-tiny-3bit-MLX | 3-bit | 3.0 GB | 6 GB | 57 |
|---|---|---|---|---|
| Granite-4.0-H-Tiny-4bit-DWQ | 4-bit DWQ | 3.9 GB | 7 GB | 144 |
| granite-4.0-h-tiny-5bit-MLX | 5-bit | 4.8 GB | 8 GB | 54 |
| granite-4.0-h-tiny-6bit-MLX | 6-bit | 5.6 GB | 9 GB | 40 |
| granite-4.0-h-tiny-8bit | 8-bit | 7.4 GB | 11 GB | 45 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.