4-bit Open on Hugging Face
Parameters
8.0B
Weights (est.)
≈ 4.5 GB
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.
Run it
With mlx-lm on an Apple silicon Mac:
pip install mlx-lmmlx_lm.generate --model mlx-community/aya-expanse-8b-4bit-mlx --prompt "Hello"
Details
- Base model
- CohereLabs/aya-expanse-8b (quantized)
- Vendor
- Cohere
- Task
- Text generation
- Quantisation
- 4-bit
- Licence
- cc-by-nc-4.0
- Downloads
- 55 in the last 30 days · 671 all time · 0 likes
- Published
- 2024-12-23 · updated 2024-12-23 (1y ago)
- Languages
- en, fr, de, es, it, pt, ja, ko, zh, ar, el, fa, pl, id, cs, he, hi, nl, ro, ru, tr, uk, vi
- Tags
- cohere
Other MLX builds of CohereLabs/aya-expanse-8b
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| aya-expanse-8b-4bit | 4-bit | ≈ 4.5 GB | 8 GB+ | 147 |
| aya-expanse-8b-4bit-mlx (this one) | 4-bit | ≈ 4.5 GB | 8 GB+ | 55 |
| aya-expanse-8b-8bit | 8-bit | ≈ 8.5 GB | 16 GB+ | 59 |