Quantization publisher
KuroTo4ka
KuroTo4ka publishes 10 quantizations across 1 models in our index, averaging 5.937 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 11 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
1
Quantizations
10
Models covered
1
Avg effective bpw
5.937
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | KuroTo4ka | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q4_K_S | 4.47 GiB | mradermacher | 4.47 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q5_K_S | 5.33 GiB | mradermacher | 5.33 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q6_K | 6.26 GiB | mradermacher | 6.26 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q5_K_M | 5.45 GiB | mradermacher | 5.45 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q8_0 | 8.11 GiB | mradermacher | 8.11 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | F16 | 15.26 GiB | mradermacher | 15.26 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q2_K | 3.06 GiB | mradermacher | 3.06 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q3_K_L | 4.13 GiB | mradermacher | 4.13 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q3_K_M | 3.84 GiB | mradermacher | 3.84 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q4_K_M | 4.68 GiB | mradermacher | 4.68 GiB | -0.0% |
| Qwen3-VL-8B-Instruct-Unredacted-MAX | Q8_0 | 8.11 GiB | prithivMLmods | 8.11 GiB | 0.0% |
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 |
|---|---|---|
| Qwen3-VL-8B-Instruct-Unredacted-MAX | 10 | 3.06 GiB |