MLX Community · First MLX build 11 Mar 2026
granite-4.0-1b-speech
MLX
apache-2.0
A 2.3B parameter model, with 6 MLX builds, from 4-bit to bf16.
- Parameters
- 2.3B From Hugging Face
- Quantisation
- 8-bit granite-4.0-1b-speech-8bit
- Memory needed
- 6 GB Estimate, unified memory
- Download
- 2.9 GB Weights, estimated
- Downloads
- 132 This build, last 30 days
- Task
- —
- MLX builds
- 6 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 6 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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/granite-4.0-1b-speech-8bitMLX builds
6 on mlx-community| granite-4.0-1b-speech-4bit | 4-bit | 2.0 GB | 5 GB | 71 |
|---|---|---|---|---|
| granite-4.0-1b-speech-mxfp4 | MXFP4 | 2.0 GB | 5 GB | 18 |
| granite-4.0-1b-speech-5bit | 5-bit | 2.2 GB | 5 GB | 53 |
| granite-4.0-1b-speech-6bit | 6-bit | 2.4 GB | 5 GB | 25 |
| granite-4.0-1b-speech-8bit | 8-bit | 2.9 GB | 6 GB | 132 |
| granite-4.0-1b-speech-bf16 | bf16 | 4.6 GB | 8 GB | 38 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.