Model comparison

diffusiongemma-26B-A4B-it vs Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP

At Q4_K_M, diffusiongemma-26B-A4B-it is the smaller download — 16,806,810,208 bytes against 36,545,830,912. At long context the gap widens: diffusiongemma-26B-A4B-it's KV cache at 32K is 1.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

diffusiongemma-26B-A4B-itQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTP
Parameters25.8B27.8B
Architecturediffusion-gemmaqwen35
Layers3064
Native context262,144262,144
Mixture of expertsyes, 128 expertsno
Quantizations published1234
Smallest quantization9.86 GiB10.12 GiB
Q4_K_M15.65 GiB34.04 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextdiffusiongemma-26B-A4B-itQwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-MTPRatio
4,0960.45 GiB0.25 GiB1.80×
8,1920.61 GiB0.50 GiB1.21×
16,3840.92 GiB1.00 GiB1.09×
32,7681.54 GiB2.00 GiB1.30×
65,5362.79 GiB4.00 GiB1.43×
131,0725.29 GiB8.00 GiB1.51×