MLX Community · First MLX build 26 Mar 2024
ilsp-Meltemi-7B-Instruct-v1
MLX
Text apache-2.0
A 7.5B parameter text generation model, with 2 MLX builds, from 4-bit to fp16.
- Parameters
- 7.5B From Hugging Face
- Quantisation
- 4-bit ilsp-Meltemi-7B-Instruct-v1-4bit
- Memory needed
- 8 GB Estimate, unified memory
- Download
- 4.6 GB Weights, estimated
- Downloads
- 25 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 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/ilsp-Meltemi-7B-Instruct-v1-4bit --prompt "Hello"MLX builds
2 on mlx-community| ilsp-Meltemi-7B-Instruct-v1-4bit | 4-bit | 4.6 GB | 8 GB | 25 |
|---|---|---|---|---|
| ilsp-Meltemi-7B-Instruct-v1-fp16 | fp16 | 4.6 GB | 8 GB | 23 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.