Is gemma-4-E2B-it-qat-q4_0-unquantized-heretic or Qwythos-9B-Claude-Mythos-5-1M smaller?

gemma-4-E2B-it-qat-q4_0-unquantized-heretic is smaller at Q4_K_M: 3.18 GiB against 10.73 GiB.

Which uses less memory at long context?

gemma-4-E2B-it-qat-q4_0-unquantized-heretic — its KV cache at 32K context is 4.1x smaller, because of how its attention layers cache.

Model comparison

gemma-4-E2B-it-qat-q4_0-unquantized-heretic vs Qwythos-9B-Claude-Mythos-5-1M

At Q4_K_M, gemma-4-E2B-it-qat-q4_0-unquantized-heretic is the smaller download — 3,416,119,296 bytes against 11,516,776,704. At long context the gap widens: gemma-4-E2B-it-qat-q4_0-unquantized-heretic's KV cache at 32K is 4.1× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-4-E2B-it-qat-q4_0-unquantized-heretic	Qwythos-9B-Claude-Mythos-5-1M
Parameters	5.1B	9.4B
Architecture	gemma4	qwen35
Layers	35	32
Native context	131,072	1,048,576
Mixture of experts	no	no
Quantizations published	28	16
Smallest quantization	2.16 GiB	5.38 GiB
Q4_K_M	3.18 GiB	10.73 GiB
Licence	gemma	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-4-E2B-it-qat-q4_0-unquantized-heretic	Qwythos-9B-Claude-Mythos-5-1M	Ratio
4,096	0.05 GiB	0.13 GiB	2.29×
8,192	0.08 GiB	0.25 GiB	3.05×
16,384	0.14 GiB	0.50 GiB	3.66×
32,768	0.25 GiB	1.00 GiB	4.06×
65,536	0.46 GiB	2.00 GiB	4.30×
131,072	0.90 GiB	4.00 GiB	4.43×