MLX Community · First MLX build 26 Jul 2024
Llama-3.1-70B-Japanese-Instruct-2407
MLX
Text llama3.1
A 71B parameter text generation model, with 2 MLX builds, from 4-bit to 8-bit.
- Parameters
- 71B From Hugging Face
- Quantisation
- 8-bit Llama-3.1-70B-Japanese-Instruct-2407-8bit
- Memory needed
- 86 GB Estimate, unified memory
- Download
- 75.0 GB Weights, estimated
- Downloads
- 197 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 86 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Llama-3.1-70B-Japanese-Instruct-2407-8bit --prompt "Hello"MLX builds
2 on mlx-community| Llama-3.1-70B-Japanese-Instruct-2407-4bit | 4-bit | 39.7 GB | 47 GB | 165 |
|---|---|---|---|---|
| Llama-3.1-70B-Japanese-Instruct-2407-8bit | 8-bit | 75.0 GB | 86 GB | 197 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.