arcee-ai · First MLX build 11 Dec 2024
Virtuoso-Small
MLX
apache-2.0
A 15B parameter model from arcee-ai, with 3 MLX builds, from 4-bit to 8-bit.
- Parameters
- 15B From Hugging Face
- Quantisation
- 6-bit Virtuoso-Small-6bit
- Memory needed
- 16 GB Estimate, unified memory
- Download
- 12.0 GB Weights, estimated
- Downloads
- 17 This build, last 30 days
- Task
- —
- MLX builds
- 3 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 16 GB at 6-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/Virtuoso-Small-6bitMLX builds
3 on mlx-community| Virtuoso-Small-4bit | 4-bit | 8.3 GB | 12 GB | 12 |
|---|---|---|---|---|
| Virtuoso-Small-6bit | 6-bit | 12.0 GB | 16 GB | 17 |
| Virtuoso-Small-8bit | 8-bit | 15.7 GB | 20 GB | 14 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.