Is gemma-4-12B-it-qat-q4_0-unquantized or gemma-4-A4B-98e-v7-coderx-it smaller?

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

Which uses less memory at long context?

gemma-4-A4B-98e-v7-coderx-it — its KV cache at 32K context is 1.6x smaller, because of how its attention layers cache.

Model comparison

gemma-4-12B-it-qat-q4_0-unquantized vs gemma-4-A4B-98e-v7-coderx-it

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-A4B-98e-v7-coderx-it's KV cache at 32K is 1.6× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-4-12B-it-qat-q4_0-unquantized	gemma-4-A4B-98e-v7-coderx-it
Parameters	12.0B	20.5B
Architecture	gemma4	gemma4
Layers	48	30
Native context	262,144	262,144
Mixture of experts	no	yes, 98 experts
Quantizations published	2	14
Smallest quantization	6.50 GiB	8.22 GiB
Q4_K_M	—	12.33 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-4-12B-it-qat-q4_0-unquantized	gemma-4-A4B-98e-v7-coderx-it	Ratio
4,096	0.72 GiB	0.45 GiB	1.60×
8,192	0.97 GiB	0.61 GiB	1.60×
16,384	1.47 GiB	0.92 GiB	1.60×
32,768	2.47 GiB	1.54 GiB	1.60×
65,536	4.47 GiB	2.79 GiB	1.60×
131,072	8.47 GiB	5.29 GiB	1.60×