IBM · First MLX build 28 Oct 2025
granite-4.0-350m-base
MLX
Text apache-2.0
A 352M parameter text generation model from IBM, with 6 MLX builds, from 3-bit to bf16.
- Parameters
- 352M From Hugging Face
- Quantisation
- 4-bit granite-4.0-350m-base-4bit
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 198 MB Weights, estimated
- Downloads
- 32 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 6 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/granite-4.0-350m-base-4bit --prompt "Hello"MLX builds
6 on mlx-community| granite-4.0-350m-base-3bit | 3-bit | 154 MB | 3 GB | 22 |
|---|---|---|---|---|
| granite-4.0-350m-base-4bit | 4-bit | 198 MB | 3 GB | 32 |
| granite-4.0-350m-base-5bit | 5-bit | 242 MB | 3 GB | 21 |
| granite-4.0-350m-base-6bit | 6-bit | 286 MB | 3 GB | 21 |
| granite-4.0-350m-base-8bit | 8-bit | 374 MB | 3 GB | 26 |
| granite-4.0-350m-base-bf16 | bf16 | 705 MB | 3 GB | 15 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.