Exllamav2 quant (exl2 / 4.0 bpw) made with ExLlamaV2 v0.1.3
Other EXL2 quants:
Quant |
Model Size |
lm_head |
2.2 |
18625 MB |
6 |
2.5 |
20645 MB |
6 |
3.0 |
24211 MB |
6 |
3.5 |
27784 MB |
6 |
3.75 |
29572 MB |
6 |
4.0 |
31359 MB |
6 |
4.25 |
33139 MB |
6 |
5.0 |
38500 MB |
6 |
6.0 |
45805 MB |
8 |
6.5 |
49410 MB |
8 |
8.0 |
54655 MB |
8 |