Quantization publisher

tencent

tencent publishes 6 quantizations across 2 models in our index, averaging 6.071 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
2
Quantizations
6
Models covered
2
Avg effective bpw
6.071
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuanttencentvsTheirsDifference
Hy-MT2-7BQ6_K5.74 GiBunsloth5.74 GiB-0.0%
Hy-MT2-7BQ8_07.43 GiBunsloth7.43 GiB-0.0%
Hy-MT2-7BQ4_K_M4.31 GiBunsloth4.31 GiB-0.0%
Hy-MT2-7BQ6_K5.74 GiBmradermacher5.74 GiB-0.0%
Hy-MT2-7BQ8_07.43 GiBmradermacher7.43 GiB-0.0%
Hy-MT2-7BQ4_K_M4.31 GiBmradermacher4.31 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
Hy-MT2-7B34.31 GiB
Hy-MT2-1.8B31.06 GiB