Is gemma-4-12B-it-qat-q4_0-unquantized or NVIDIA-Nemotron-Nano-9B-v2 smaller?

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

Which uses less memory at long context?

gemma-4-12B-it-qat-q4_0-unquantized — its KV cache at 32K context is 2.8x smaller, because of how its attention layers cache.

Model comparison

gemma-4-12B-it-qat-q4_0-unquantized vs NVIDIA-Nemotron-Nano-9B-v2

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-12B-it-qat-q4_0-unquantized's KV cache at 32K is 2.8× 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	NVIDIA-Nemotron-Nano-9B-v2
Parameters	12.0B	8.9B
Architecture	gemma4	nemotron_h
Layers	48	56
Native context	262,144	131,072
Mixture of experts	no	no
Quantizations published	2	34
Smallest quantization	6.50 GiB	4.62 GiB
Q4_K_M	—	6.08 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	gemma-4-12B-it-qat-q4_0-unquantized	NVIDIA-Nemotron-Nano-9B-v2	Ratio
4,096	0.72 GiB	0.88 GiB	1.22×
8,192	0.97 GiB	1.75 GiB	1.81×
16,384	1.47 GiB	3.50 GiB	2.38×
32,768	2.47 GiB	7.00 GiB	2.84×
65,536	4.47 GiB	14.00 GiB	3.13×
131,072	8.47 GiB	28.00 GiB	3.31×