Quantization publisher

0xgr3y

0xgr3y publishes 5 quantizations across 1 models in our index, averaging 8.658 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
5
Models covered
1
Avg effective bpw
8.658
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuant0xgr3yvsTheirsDifference
Qwen2.5-Coder-0.5BQ4_K_M0.37 GiBbartowski0.37 GiB-0.0%
Qwen2.5-Coder-0.5BQ5_K_M0.39 GiBbartowski0.39 GiB-0.0%
Qwen2.5-Coder-0.5BQ6_K0.47 GiBbartowski0.47 GiB-0.0%
Qwen2.5-Coder-0.5BF160.93 GiBbartowski0.93 GiB-0.0%
Qwen2.5-Coder-0.5BQ4_K_M0.37 GiBlmstudio-community0.37 GiB-0.0%
Qwen2.5-Coder-0.5BQ6_K0.47 GiBlmstudio-community0.47 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
Qwen2.5-Coder-0.5B50.33 GiB