IBM · First MLX build 16 Apr 2025
granite-3.3-8b-instruct
MLX
Text apache-2.0
A 8.2B parameter text generation model from IBM, with 4 MLX builds, from 4-bit to fp16.
- Parameters
- 8.2B From Hugging Face
- Quantisation
- 4-bit granite-3.3-8b-instruct-4bit
- Memory needed
- 8 GB Estimate, unified memory
- Download
- 4.6 GB Weights, estimated
- Downloads
- 222 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 4 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 8 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-3.3-8b-instruct-4bit --prompt "Hello"MLX builds
4 on mlx-community| granite-3.3-8b-instruct-4bit | 4-bit | 4.6 GB | 8 GB | 222 |
|---|---|---|---|---|
| granite-3.3-8b-instruct-6bit | 6-bit | 6.6 GB | 10 GB | 27 |
| granite-3.3-8b-instruct-8bit | 8-bit | 8.7 GB | 12 GB | 44 |
| granite-3.3-8b-instruct-fp16 | fp16 | 16.3 GB | 21 GB | 53 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.