mamba2-370m-4bit
mlx-community/mamba2-370m-4bit
Uncategorised · mlx
4-bit Open on Hugging Face
Parameters
58M
Weights (est.)
≈ 50 MB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Details
- Base model
- rokyang/mamba2-370m-hf (finetune)
- Vendor
- rokyang
- Task
- Uncategorised
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 19 in the last 30 days · 209 all time · 0 likes
- Published
- 2025-01-21 · updated 2025-01-21 (1y ago)
Other MLX builds of rokyang/mamba2-370m-hf
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| mamba2-370m-4bit (this one) | 4-bit | ≈ 50 MB | 8 GB+ | 19 |
| mamba2-370m-8bit | 8-bit | ≈ 122 MB | 8 GB+ | 21 |
| mamba2-370m | — | ≈ 737 MB | 8 GB+ | 18 |