Is gemma-4-12B-it-qat-q4_0-unquantized or SOLAR-10.7B-Instruct-v1.0-uncensored 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 2.4x smaller, because of how its attention layers cache.

Model comparison

gemma-4-12B-it-qat-q4_0-unquantized vs SOLAR-10.7B-Instruct-v1.0-uncensored

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 2.4× 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	SOLAR-10.7B-Instruct-v1.0-uncensored
Parameters	12.0B	10.7B
Architecture	gemma4	llama
Layers	48	48
Native context	262,144	4,096
Mixture of experts	no	no
Quantizations published	2	12
Smallest quantization	6.50 GiB	4.24 GiB
Q4_K_M	—	6.02 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	SOLAR-10.7B-Instruct-v1.0-uncensored	Ratio
4,096	0.72 GiB	0.75 GiB	1.04×
8,192	0.97 GiB	1.50 GiB	1.55×
16,384	1.47 GiB	3.00 GiB	2.04×
32,768	2.47 GiB	6.00 GiB	2.43×
65,536	4.47 GiB	12.00 GiB	2.69×
131,072	8.47 GiB	24.00 GiB	2.83×