Is medgemma-27b-it or Qwen3.5-9B smaller?

Qwen3.5-9B is smaller at Q4_K_M: 5.24 GiB against 15.41 GiB.

Which uses less memory at long context?

Qwen3.5-9B — its KV cache at 32K context is 3.1x smaller, because of how its attention layers cache.

Model comparison

medgemma-27b-it vs Qwen3.5-9B

At Q4_K_M, Qwen3.5-9B is the smaller download — 5,627,044,256 bytes against 16,546,689,536. At long context the gap widens: Qwen3.5-9B's KV cache at 32K is 3.1× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	medgemma-27b-it	Qwen3.5-9B
Parameters	28.8B	9.7B
Architecture	gemma3	qwen35
Layers	62	32
Native context	131,072	262,144
Mixture of experts	no	no
Quantizations published	68	34
Smallest quantization	5.83 GiB	2.97 GiB
Q4_K_M	15.41 GiB	5.24 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	medgemma-27b-it	Qwen3.5-9B	Ratio
4,096	0.92 GiB	0.13 GiB	7.37×
8,192	1.23 GiB	0.25 GiB	4.94×
16,384	1.86 GiB	0.50 GiB	3.72×
32,768	3.11 GiB	1.00 GiB	3.11×
65,536	5.61 GiB	2.00 GiB	2.80×
131,072	10.61 GiB	4.00 GiB	2.65×