Mistral AI · First MLX build 20 Mar 2026
Leanstral-2603
MLX
apache-2.0
A 119B parameter model from Mistral AI, with one MLX build at 8-bit.
- Parameters
- 119B From Hugging Face
- Quantisation
- 8-bit Leanstral-2603-8bit
- Memory needed
- 145 GB Estimate, unified memory
- Download
- 127 GB Weights, estimated
- Downloads
- 26 This build, last 30 days
- Task
- —
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 145 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/Leanstral-2603-8bitMLX builds
1 on mlx-community| Leanstral-2603-8bit | 8-bit | 127 GB | 145 GB | 26 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.