Model comparison

gemma4-26B-it vs Qwen3.6-35B-A3B

At Q4_K_M, gemma4-26B-it is the smaller download — 16,796,015,488 bytes against 21,166,757,728.

From the file· summed bytes, KV per layer

Side by side

gemma4-26B-itQwen3.6-35B-A3B
Parameters25.2B36.0B
Architecturegemma4qwen35moe
Layers40
Native context262,144
Mixture of expertsnoyes, 256 experts
Quantizations published461
Smallest quantization15.64 GiB8.77 GiB
Q4_K_M15.64 GiB19.71 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextgemma4-26B-itQwen3.6-35B-A3BRatio
4,0960.08 GiB
8,1920.16 GiB
16,3840.31 GiB
32,7680.63 GiB
65,5361.25 GiB
131,0722.50 GiB