Is command-r-35b-writer-v2 or gemma-4-12B-it-qat-q4_0-unquantized 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 16.2x smaller, because of how its attention layers cache.

Model comparison

command-r-35b-writer-v2 vs gemma-4-12B-it-qat-q4_0-unquantized

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 16.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	command-r-35b-writer-v2	gemma-4-12B-it-qat-q4_0-unquantized
Parameters	35.0B	12.0B
Architecture	command-r	gemma4
Layers	40	48
Native context	8,192	262,144
Mixture of experts	no	no
Quantizations published	34	2
Smallest quantization	7.94 GiB	6.50 GiB
Q4_K_M	20.05 GiB	—
Licence	cc-by-nc-4.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	command-r-35b-writer-v2	gemma-4-12B-it-qat-q4_0-unquantized	Ratio
4,096	5.00 GiB	0.72 GiB	6.96×
8,192	10.00 GiB	0.97 GiB	10.32×
16,384	20.00 GiB	1.47 GiB	13.62×
32,768	40.00 GiB	2.47 GiB	16.20×
65,536	80.00 GiB	4.47 GiB	17.90×
131,072	160.00 GiB	8.47 GiB	18.89×