Is gemma-3n-E2B-it or Qwythos-9B-v2 smaller?

gemma-3n-E2B-it is smaller at Q4_K_M: 2.60 GiB against 5.63 GiB.

Which uses less memory at long context?

gemma-3n-E2B-it — its KV cache at 32K context is 2.4x smaller, because of how its attention layers cache.

Model comparison

gemma-3n-E2B-it vs Qwythos-9B-v2

At Q4_K_M, gemma-3n-E2B-it is the smaller download — 2,787,806,304 bytes against 6,047,707,680. At long context the gap widens: gemma-3n-E2B-it's KV cache at 32K is 2.4× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-3n-E2B-it	Qwythos-9B-v2
Parameters	5.4B	9.7B
Architecture	gemma3n	qwen35
Layers	30	32
Native context	32,768	1,048,576
Mixture of experts	no	no
Quantizations published	40	43
Smallest quantization	1.76 GiB	3.64 GiB
Q4_K_M	2.60 GiB	5.63 GiB
Licence	gemma	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-3n-E2B-it	Qwythos-9B-v2	Ratio
4,096	0.09 GiB	0.13 GiB	1.33×
8,192	0.14 GiB	0.25 GiB	1.78×
16,384	0.23 GiB	0.50 GiB	2.13×
32,768	0.42 GiB	1.00 GiB	2.37×
65,536	0.80 GiB	2.00 GiB	2.51×
131,072	1.55 GiB	4.00 GiB	2.59×