Is gemma-4-12B-it-qat-q4_0-unquantized or Trinity-Mini smaller?

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

Which uses less memory at long context?

Trinity-Mini — its KV cache at 32K context is 4.0x smaller, because of how its attention layers cache.

Model comparison

gemma-4-12B-it-qat-q4_0-unquantized vs Trinity-Mini

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: Trinity-Mini's KV cache at 32K is 4.0× 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	Trinity-Mini
Parameters	12.0B	26.1B
Architecture	gemma4	afmoe
Layers	48	32
Native context	262,144	131,072
Mixture of experts	no	yes, 128 experts
Quantizations published	2	57
Smallest quantization	6.50 GiB	6.11 GiB
Q4_K_M	—	14.74 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	Trinity-Mini	Ratio
4,096	0.72 GiB	0.18 GiB	4.00×
8,192	0.97 GiB	0.24 GiB	4.00×
16,384	1.47 GiB	0.37 GiB	4.00×
32,768	2.47 GiB	0.62 GiB	4.00×
65,536	4.47 GiB	1.12 GiB	4.00×
131,072	8.47 GiB	2.12 GiB	4.00×