Google · First MLX build 15 Jun 2026
gemma-4-12B-it-qat-q4_0-unquantized
MLX
TextVisionAudio apache-2.0
A 12B parameter any to any model from Google, with one MLX build at 5-bit.
- Parameters
- 12B From Hugging Face
- Quantisation
- 5-bit Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-5bit-msq
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 3.7 GB Weights, estimated
- Downloads
- 187 This build, last 30 days
- Task
- Any to any Text, Vision, Audio
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 7 GB at 5-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/Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-5bit-msqMLX builds
1 on mlx-community| Huihui-gemma-4-12B-it-qat-q4_0-unquantized-abliterated-5bit-msq | 5-bit | 3.7 GB | 7 GB | 187 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.