Quantization publisher
Majipa
Majipa publishes 20 quantizations across 1 models in our index, averaging 6.394 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 3 of the pairs we can compare — the same label does not mean the same file.
From the file· summed file bytes
Repositories
1
Quantizations
20
Models covered
1
Avg effective bpw
6.394
across their files
Same model, same quant label, different bytes
largest disagreements first
| Model | Quant | Majipa | vs | Theirs | Difference |
|---|---|---|---|---|---|
| Qwen3-0.6B-Base | Q8_0 | 0.75 GiB | Qwen | 0.60 GiB | +25.9% |
| Qwen3-0.6B-Base | Q8_0 | 0.75 GiB | iyanello | 0.60 GiB | +25.9% |
| Qwen3-0.6B-Base | Q4_K_M | 0.45 GiB | iyanello | 0.37 GiB | +22.1% |
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 |
|---|---|---|
| Qwen3-0.6B-Base | 20 | 0.29 GiB |