Quantization publisher

leeminwaan

leeminwaan publishes 4 quantizations across 1 models in our index, averaging 5.377 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 6 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
1
Quantizations
4
Models covered
1
Avg effective bpw
5.377
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantleeminwaanvsTheirsDifference
Qwen3.5-4BQ5_K_M2.86 GiBlainlives2.90 GiB-1.1%
Qwen3.5-4BQ4_K_M2.52 GiBlainlives2.52 GiB0.0%
Qwen3.5-4BQ8_04.17 GiBlainlives4.17 GiB-0.0%
Qwen3.5-4BQ4_K_M2.52 GiBJackrong2.52 GiB0.0%
Qwen3.5-4BQ5_K_M2.86 GiBJackrong2.86 GiB0.0%
Qwen3.5-4BQ8_04.17 GiBJackrong4.17 GiB0.0%

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

ModelQuantizationsSmallest
Qwen3.5-4B42.11 GiB