MLX Community · First MLX build 5 May 2024
Meta-Llama-3-120B-Instruct
MLX
other
A 122B parameter model, with one MLX build at 4-bit.
- Parameters
- 122B From Hugging Face
- Quantisation
- 4-bit Meta-Llama-3-120B-Instruct-4bit
- Memory needed
- 79 GB Estimate, unified memory
- Download
- 68.6 GB Weights, estimated
- Downloads
- 25 This build, last 30 days
- Task
- —
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 79 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/Meta-Llama-3-120B-Instruct-4bitMLX builds
1 on mlx-community| Meta-Llama-3-120B-Instruct-4bit | 4-bit | 68.6 GB | 79 GB | 25 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.