Is gemma-4-E2B-it-qat-q4_0-unquantized-heretic or gemma-4-E2B-it smaller?

gemma-4-E2B-it is smaller at Q4_K_M: 2.89 GiB against 3.18 GiB.

Which uses less memory at long context?

gemma-4-E2B-it — its KV cache at 32K context is 1.0x smaller, because of how its attention layers cache.

Model comparison

At Q4_K_M, gemma-4-E2B-it is the smaller download — 3,106,738,272 bytes against 3,416,119,296.

From the file· summed bytes, KV per layer

	gemma-4-E2B-it-qat-q4_0-unquantized-heretic	gemma-4-E2B-it
Parameters	5.1B	5.1B
Architecture	gemma4	gemma4
Layers	35	35
Native context	131,072	131,072
Mixture of experts	no	no
Quantizations published	28	54
Smallest quantization	2.16 GiB	2.13 GiB
Q4_K_M	3.18 GiB	2.89 GiB
Licence	gemma	apache-2.0

the term that decides long-context viability

Context	gemma-4-E2B-it-qat-q4_0-unquantized-heretic	gemma-4-E2B-it	Ratio
4,096	0.05 GiB	0.05 GiB	1.00×
8,192	0.08 GiB	0.08 GiB	1.00×
16,384	0.14 GiB	0.14 GiB	1.00×
32,768	0.25 GiB	0.25 GiB	1.00×
65,536	0.46 GiB	0.46 GiB	1.00×
131,072	0.90 GiB	0.90 GiB	1.00×