Is medgemma-1.5-4b-it or Qwen3-VL-4B-Instruct smaller?

medgemma-1.5-4b-it is smaller at Q4_K_M: 2.32 GiB against 2.33 GiB.

Which uses less memory at long context?

medgemma-1.5-4b-it — its KV cache at 32K context is 5.7x smaller, because of how its attention layers cache.

Model comparison

medgemma-1.5-4b-it vs Qwen3-VL-4B-Instruct

At Q4_K_M, medgemma-1.5-4b-it is the smaller download — 2,489,894,624 bytes against 2,497,281,568. At long context the gap widens: medgemma-1.5-4b-it's KV cache at 32K is 5.7× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	medgemma-1.5-4b-it	Qwen3-VL-4B-Instruct
Parameters	4.3B	4.4B
Architecture	gemma3	qwen3vl
Layers	34	36
Native context	131,072	262,144
Mixture of experts	no	no
Quantizations published	32	26
Smallest quantization	1.10 GiB	1.01 GiB
Q4_K_M	2.32 GiB	2.33 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	medgemma-1.5-4b-it	Qwen3-VL-4B-Instruct	Ratio
4,096	0.25 GiB	0.56 GiB	2.27×
8,192	0.33 GiB	1.13 GiB	3.45×
16,384	0.48 GiB	2.25 GiB	4.66×
32,768	0.79 GiB	4.50 GiB	5.66×
65,536	1.42 GiB	9.00 GiB	6.34×
131,072	2.67 GiB	18.00 GiB	6.74×