Model comparison

gemma-4-26B-A4B-it vs gemma-4-26B-A4B-it-qat-q4_0-unquantized

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

gemma-4-26B-A4B-itgemma-4-26B-A4B-it-qat-q4_0-unquantized
Parameters26.5B26.5B
Architecturegemma4gemma4
Layers3030
Native context262,144262,144
Mixture of expertsyes, 128 expertsyes, 128 experts
Quantizations published461
Smallest quantization8.99 GiB13.45 GiB
Q4_K_M15.64 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextgemma-4-26B-A4B-itgemma-4-26B-A4B-it-qat-q4_0-unquantizedRatio
4,0960.45 GiB0.45 GiB1.00×
8,1920.61 GiB0.61 GiB1.00×
16,3840.92 GiB0.92 GiB1.00×
32,7681.54 GiB1.54 GiB1.00×
65,5362.79 GiB2.79 GiB1.00×
131,0725.29 GiB5.29 GiB1.00×