Quantization publisher
hans00
hans00 publishes 17 quantizations across 1 models in our index, averaging 12.337 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 5 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
1
Quantizations
17
Models covered
1
Avg effective bpw
12.337
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | hans00 | vs | Theirs | Difference |
|---|---|---|---|---|---|
| chatterbox | BF16 | 0.94 GiB | calcuis | 1.48 GiB | -36.9% |
| chatterbox | F32 | 2.44 GiB | calcuis | 2.97 GiB | -18.0% |
| chatterbox | Q5_K_M | 0.52 GiB | calcuis | 0.78 GiB | -33.8% |
| chatterbox | F16 | 1.25 GiB | calcuis | 1.49 GiB | -15.9% |
| chatterbox | Q4_K_M | 0.46 GiB | calcuis | 0.66 GiB | -31.1% |
A quantization label describes a target, not a recipe. Publishers make different choices about which tensors to keep at higher precision, and some apply an importance matrix while others don't — so two files both honestly labelled the same thing can differ measurably in size and in quality.
Models they publish
| Model | Quantizations | Smallest |
|---|---|---|
| chatterbox | 17 | 0.17 GiB |