Quantization publisher

Rangle2

Rangle2 publishes 7 quantizations across 1 models in our index, averaging 6.989 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
7
Models covered
1
Avg effective bpw
6.989
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantRangle2vsTheirsDifference
gemma-4-12B-it-uncensored-opus4.7-cotQ2_K4.50 GiBmradermacher4.50 GiB-0.0%
gemma-4-12B-it-uncensored-opus4.7-cotQ3_K_M5.67 GiBmradermacher5.67 GiB-0.0%
gemma-4-12B-it-uncensored-opus4.7-cotQ4_K_M6.87 GiBmradermacher6.87 GiB-0.0%
gemma-4-12B-it-uncensored-opus4.7-cotQ5_K_M7.96 GiBmradermacher7.96 GiB-0.0%
gemma-4-12B-it-uncensored-opus4.7-cotQ6_K9.11 GiBmradermacher9.11 GiB-0.0%
gemma-4-12B-it-uncensored-opus4.7-cotQ8_011.80 GiBmradermacher11.80 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
gemma-4-12B-it-uncensored-opus4.7-cot74.50 GiB