SorcererLM-8x22b-8bit
mlx-community/SorcererLM-8x22b-8bit
Uncategorised · mlx
8-bit Open on Hugging Face
Parameters
141B
Weights (est.)
≈ 149 GB
Mac memory (est.)
192 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
- rAIfle/SorcererLM-8x22b-bf16 (quantized)
- Vendor
- rAIfle
- Task
- Uncategorised
- Quantisation
- 8-bit
- Licence
- apache-2.0
- Downloads
- 14 in the last 30 days · 223 all time · 0 likes
- Published
- 2024-10-13 · updated 2024-10-14 (2y ago)
- Tags
- mixtral
Other MLX builds of rAIfle/SorcererLM-8x22b-bf16
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| SorcererLM-8x22b-2bit | 2-bit | ≈ 43.9 GB | 64 GB+ | 16 |
| SorcererLM-8x22b-4bit | 4-bit | ≈ 79.1 GB | 96 GB+ | 21 |
| SorcererLM-8x22b-8bit (this one) | 8-bit | ≈ 149 GB | 192 GB+ | 14 |