Is Qwen3.5-9B or LFM2-VL-3B smaller?

LFM2-VL-3B is smaller at Q4_K_M: 1.46 GiB against 5.24 GiB.

Which uses less memory at long context?

LFM2-VL-3B — its KV cache at 32K context is 2.0x smaller, because of how its attention layers cache.

Model comparison

Qwen3.5-9B vs LFM2-VL-3B

At Q4_K_M, LFM2-VL-3B is the smaller download — 1,563,668,256 bytes against 5,627,044,256. At long context the gap widens: LFM2-VL-3B's KV cache at 32K is 2.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.5-9B	LFM2-VL-3B
Parameters	9.7B	3.0B
Architecture	qwen35	lfm2
Layers	32	30
Native context	262,144	128,000
Mixture of experts	no	no
Quantizations published	34	7
Smallest quantization	2.97 GiB	1.38 GiB
Q4_K_M	5.24 GiB	1.46 GiB
Licence	apache-2.0	other

KV cache by context

the term that decides long-context viability

Context	Qwen3.5-9B	LFM2-VL-3B	Ratio
4,096	0.13 GiB	0.06 GiB	2.00×
8,192	0.25 GiB	0.13 GiB	2.00×
16,384	0.50 GiB	0.25 GiB	2.00×
32,768	1.00 GiB	0.50 GiB	2.00×
65,536	2.00 GiB	1.00 GiB	2.00×
131,072	4.00 GiB	2.00 GiB	2.00×