Quantization publisher

xthor

xthor publishes 3 quantizations across 1 models in our index, averaging 9.996 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 4 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
1
Quantizations
3
Models covered
1
Avg effective bpw
9.996
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantxthorvsTheirsDifference
Qwen3-Embedding-0.6BF161.12 GiBmradermacher1.12 GiB-0.0%
Qwen3-Embedding-0.6BQ4_K_M0.37 GiBmradermacher0.37 GiB-0.0%
Qwen3-Embedding-0.6BQ8_00.60 GiBmradermacher0.60 GiB-0.0%
Qwen3-Embedding-0.6BQ8_00.60 GiBPeterAM40.60 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-Embedding-0.6B30.37 GiB