Is Qwen3-Coder-30B-A3B-Instruct-RTPurbo 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 1.2x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct-RTPurbo 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 1.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3-Coder-30B-A3B-Instruct-RTPurbo	gemma-4-12B-it-qat-q4_0-unquantized
Parameters	30.5B	12.0B
Architecture	qwen3moe	gemma4
Layers	48	48
Native context	262,144	262,144
Mixture of experts	yes, 128 experts	no
Quantizations published	23	2
Smallest quantization	5.98 GiB	6.50 GiB
Q4_K_M	—	—
Licence	—	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct-RTPurbo	gemma-4-12B-it-qat-q4_0-unquantized	Ratio
4,096	0.38 GiB	0.72 GiB	1.92×
8,192	0.75 GiB	0.97 GiB	1.29×
16,384	1.50 GiB	1.47 GiB	1.02×
32,768	3.00 GiB	2.47 GiB	1.22×
65,536	6.00 GiB	4.47 GiB	1.34×
131,072	12.00 GiB	8.47 GiB	1.42×