Google · First MLX build 12 Mar 2025
gemma-3-1b-it
MLX
Text gemma
A 1.3B parameter text generation model from Google, with 10 MLX builds, from 4-bit to bf16.
- Parameters
- 1.3B From Hugging Face
- Quantisation
- 4-bit gemma-3-1b-it-qat-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 732 MB Weights, estimated
- Downloads
- 23K This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 10 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 3 GB at 4-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-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/gemma-3-1b-it-qat-4bit --prompt "Hello"MLX builds
10 on mlx-community| gemma-3-1b-it-qat-4bit | 4-bit | 732 MB | 3 GB | 23K |
|---|---|---|---|---|
| gemma-3-1b-it-4bit | 4-bit | 732 MB | 3 GB | 7.3K |
| gemma-3-1b-it-4bit-DWQ | 4-bit DWQ | 732 MB | 3 GB | 879 |
| gemma-3-1b-it-DQ | 4-bit | 732 MB | 3 GB | 45 |
| gemma-3-1b-it-6bit | 6-bit | 1.1 GB | 4 GB | 53 |
| gemma-3-1b-it-qat-6bit | 6-bit | 1.1 GB | 4 GB | 31 |
| gemma-3-1b-it-8bit | 8-bit | 1.4 GB | 4 GB | 143 |
| gemma-3-1b-it-qat-8bit | 8-bit | 1.4 GB | 4 GB | 100 |
| gemma-3-1b-it-bf16 | bf16 | 2.6 GB | 5 GB | 736 |
| gemma-3-1b-it-qat-bf16 | bf16 | 2.6 GB | 5 GB | 37 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.