Is gemma-4-31B-it-qat-q4_0-unquantized-heretic or gemma-4-E4B-it smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

gemma-4-E4B-it — its KV cache at 32K context is 12.2x smaller, because of how its attention layers cache.

Model comparison

gemma-4-31B-it-qat-q4_0-unquantized-heretic vs gemma-4-E4B-it

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 12.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-4-31B-it-qat-q4_0-unquantized-heretic	gemma-4-E4B-it
Parameters	31.3B	8.0B
Architecture	gemma4	gemma4
Layers	60	42
Native context	262,144	131,072
Mixture of experts	no	no
Quantizations published	23	36
Smallest quantization	6.67 GiB	3.30 GiB
Q4_K_M	—	4.64 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-4-31B-it-qat-q4_0-unquantized-heretic	gemma-4-E4B-it	Ratio
4,096	1.80 GiB	0.12 GiB	14.60×
8,192	2.42 GiB	0.18 GiB	13.63×
16,384	3.67 GiB	0.29 GiB	12.79×
32,768	6.17 GiB	0.51 GiB	12.20×
65,536	11.17 GiB	0.94 GiB	11.84×
131,072	21.17 GiB	1.82 GiB	11.64×