Quantization publisher

Azure99

Azure99 publishes 8 quantizations across 1 models in our index, averaging 6.795 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
8
Models covered
1
Avg effective bpw
6.795
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantAzure99vsTheirsDifference
Qwen3.5-35B-A3B-BaseQ5_K_M23.03 GiBmradermacher23.06 GiB-0.1%
Qwen3.5-35B-A3B-BaseQ4_K_S18.52 GiBmradermacher18.52 GiB0.0%
Qwen3.5-35B-A3B-BaseQ4_K_M19.71 GiBmradermacher19.72 GiB-0.0%
Qwen3.5-35B-A3B-BaseQ5_K_S22.33 GiBmradermacher22.33 GiB-0.0%
Qwen3.5-35B-A3B-BaseQ6_K26.56 GiBmradermacher26.56 GiB-0.0%
Qwen3.5-35B-A3B-BaseQ8_034.37 GiBmradermacher34.37 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

ModelQuantizationsSmallest
Qwen3.5-35B-A3B-Base818.36 GiB