state-spaces · First MLX build 21 Sept 2024
mamba-370m-hf
MLX
Text
A 372M parameter text generation model from state-spaces, with 2 MLX builds.
- Parameters
- 372M From Hugging Face
- Quantisation
- Unstated mamba-370m-hf-f16
- Memory needed
- 3 GB Estimate, unified memory
- Download
- 743 MB Weights, estimated
- Downloads
- 21 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 3 GB at Unstated
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/mamba-370m-hf-f16 --prompt "Hello"MLX builds
2 on mlx-community| mamba-370m-hf-f16 | Unstated | 743 MB | 3 GB | 21 |
|---|---|---|---|---|
| mamba-370m-hf-f32 | Unstated | 1.5 GB | 4 GB | 18 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.