Model comparison

Fallen-Gemma3-27B-v1 vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 16,546,688,704. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 4.6× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Fallen-Gemma3-27B-v1gemma-4-E4B-it
Parameters27.4B8.0B
Architecturegemma3gemma4
Layers4842
Native context131,072
Mixture of expertsnono
Quantizations published2536
Smallest quantization7.86 GiB3.30 GiB
Q4_K_M15.41 GiB4.64 GiB
Licenceotherapache-2.0

KV cache by context

the term that decides long-context viability
ContextFallen-Gemma3-27B-v1gemma-4-E4B-itRatio
4,0960.67 GiB0.12 GiB5.48×
8,1920.91 GiB0.18 GiB5.11×
16,3841.38 GiB0.29 GiB4.80×
32,7682.31 GiB0.51 GiB4.58×
65,5364.19 GiB0.94 GiB4.44×
131,0727.94 GiB1.82 GiB4.37×