Quantization publisher

RachidAR

RachidAR publishes 12 quantizations across 4 models in our index, averaging 9.446 effective bits per weight. Their files differ in size from other publishers' builds of the same nominal quantization on 7 of the pairs we can compare — the same label does not mean the same file.

From the file· summed file bytes
Repositories
4
Quantizations
12
Models covered
4
Avg effective bpw
9.446
across their files

Same model, same quant label, different bytes

largest disagreements first
ModelQuantRachidARvsTheirsDifference
gemma-4-26B-A4B-it-assistantQ4_K_M0.31 GiBAtomicChat0.30 GiB+1.3%
gemma-4-26B-A4B-it-assistantQ5_K_M0.32 GiBAtomicChat0.32 GiB+0.7%
gemma-4-26B-A4B-it-assistantQ8_00.43 GiBRadamanthys110.43 GiB0.0%
gemma-4-31B-it-qat-q4_0-unquantized-assistantBF160.89 GiBGaboo0.89 GiB0.0%
gemma-4-12B-it-qat-q4_0-unquantized-assistantQ8_00.43 GiBJanvitos0.43 GiB-0.0%
gemma-4-26B-A4B-it-assistantQ8_00.43 GiBAtomicChat0.43 GiB-0.0%
gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistantQ8_00.43 GiBJanvitos0.43 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