Is gemma-3n-E2B-it or gemma-4-E2B-it smaller?

gemma-3n-E2B-it is smaller at Q4_K_M: 2.60 GiB against 2.89 GiB.

Which uses less memory at long context?

gemma-4-E2B-it — its KV cache at 32K context is 1.7x smaller, because of how its attention layers cache.

Model comparison

gemma-3n-E2B-it vs gemma-4-E2B-it

At Q4_K_M, gemma-3n-E2B-it is the smaller download — 2,787,806,304 bytes against 3,106,738,272. At long context the gap widens: gemma-4-E2B-it's KV cache at 32K is 1.7× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-3n-E2B-it	gemma-4-E2B-it
Parameters	5.4B	5.1B
Architecture	gemma3n	gemma4
Layers	30	35
Native context	32,768	131,072
Mixture of experts	no	no
Quantizations published	40	54
Smallest quantization	1.76 GiB	2.13 GiB
Q4_K_M	2.60 GiB	2.89 GiB
Licence	gemma	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-3n-E2B-it	gemma-4-E2B-it	Ratio
4,096	0.09 GiB	0.05 GiB	1.71×
8,192	0.14 GiB	0.08 GiB	1.71×
16,384	0.23 GiB	0.14 GiB	1.71×
32,768	0.42 GiB	0.25 GiB	1.71×
65,536	0.80 GiB	0.46 GiB	1.71×
131,072	1.55 GiB	0.90 GiB	1.71×