Quantization publisher
Tinman-Lab
Tinman-Lab publishes 3 quantizations across 1 models in our index, averaging 9.476 effective bits per weight.
From the file· summed file bytes
Repositories
1
Quantizations
3
Models covered
1
Avg effective bpw
9.476
across their files
Models they publish
| Model | Quantizations | Smallest |
|---|---|---|
| Tinman-gemma4-companion-merged | 3 | 4.94 GiB |