Quantization publisher
tvall43
tvall43 publishes 17 quantizations across 4 models in our index, averaging 6.727 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 4 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
4
Quantizations
17
Models covered
4
Avg effective bpw
6.727
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | tvall43 | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Qwen3.6-35B-A3B-heretic | Q3_K_M | 15.61 GiB | Abiray | 15.61 GiB | 0.0% |
| Qwen3.6-35B-A3B-heretic | Q4_K_M | 19.71 GiB | Abiray | 19.71 GiB | 0.0% |
| Qwen3.6-35B-A3B-heretic | Q5_K_M | 23.03 GiB | Abiray | 23.03 GiB | 0.0% |
| Qwen3.6-35B-A3B-heretic | Q8_0 | 34.37 GiB | Abiray | 34.37 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 |
|---|---|---|
| Qwen3.6-14B-A3B-FableVibes | 6 | 4.96 GiB |
| Qwen3.6-14B-A3B-VibeForged-v2 | 5 | 4.96 GiB |
| Qwen3.5-14B-A3B-Claude-4.6-Opus-Reasoning-Distilled-reap | 1 | 26.36 GiB |
| Qwen3.6-35B-A3B-heretic | 5 | 15.61 GiB |