MLX Community · First MLX build 22 May 2024
Mistral-7B-Instruct-v0.3
MLX
apache-2.0
A 7.2B parameter model, with 3 MLX builds, from 4-bit to 8-bit.
- Parameters
- 7.2B From Hugging Face
- Quantisation
- 4-bit Mistral-7B-Instruct-v0.3-4bit
- Memory needed
- 7 GB Estimate, unified memory
- Download
- 4.1 GB Weights, estimated
- Downloads
- 16K This build, last 30 days
- Task
- —
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 7 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 · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/Mistral-7B-Instruct-v0.3-4bitMLX builds
3 on mlx-community| Mistral-7B-Instruct-v0.3-4bit | 4-bit | 4.1 GB | 7 GB | 16K |
|---|---|---|---|---|
| Mistral-7B-Instruct-v0.3-8bit | 8-bit | 7.7 GB | 11 GB | 134 |
| Mistral-7B-Instruct-v0.3 | Unstated | 14.5 GB | 19 GB | 650 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.