Is gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic or Qwen3-Coder-30B-A3B-Instruct smaller?

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

Which uses less memory at long context?

gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic — its KV cache at 32K context is 1.9x smaller, because of how its attention layers cache.

Model comparison

gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic vs Qwen3-Coder-30B-A3B-Instruct

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic's KV cache at 32K is 1.9× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic	Qwen3-Coder-30B-A3B-Instruct
Parameters	25.8B	30.5B
Architecture	gemma4	qwen3moe
Layers	30	48
Native context	262,144	262,144
Mixture of experts	yes, 128 experts	yes, 128 experts
Quantizations published	21	46
Smallest quantization	7.72 GiB	7.46 GiB
Q4_K_M	—	17.28 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic	Qwen3-Coder-30B-A3B-Instruct	Ratio
4,096	0.45 GiB	0.38 GiB	1.20×
8,192	0.61 GiB	0.75 GiB	1.24×
16,384	0.92 GiB	1.50 GiB	1.63×
32,768	1.54 GiB	3.00 GiB	1.94×
65,536	2.79 GiB	6.00 GiB	2.15×
131,072	5.29 GiB	12.00 GiB	2.27×