Is Qwen3.5-4B or GLM-4.1V-9B-Thinking smaller?

Qwen3.5-4B is smaller at Q4_K_M: 2.52 GiB against 5.74 GiB.

Which uses less memory at long context?

Qwen3.5-4B — its KV cache at 32K context is 1.2x smaller, because of how its attention layers cache.

Model comparison

Qwen3.5-4B vs GLM-4.1V-9B-Thinking

At Q4_K_M, Qwen3.5-4B is the smaller download — 2,707,513,696 bytes against 6,166,574,432. At long context the gap widens: Qwen3.5-4B's KV cache at 32K is 1.2× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.5-4B	GLM-4.1V-9B-Thinking
Parameters	4.7B	10.3B
Architecture	qwen35	glm4
Layers	32	40
Native context	262,144	65,536
Mixture of experts	no	no
Quantizations published	57	43
Smallest quantization	1.42 GiB	2.98 GiB
Q4_K_M	2.52 GiB	5.74 GiB
Licence	apache-2.0	mit

KV cache by context

the term that decides long-context viability

Context	Qwen3.5-4B	GLM-4.1V-9B-Thinking	Ratio
4,096	0.13 GiB	0.16 GiB	1.25×
8,192	0.25 GiB	0.31 GiB	1.25×
16,384	0.50 GiB	0.63 GiB	1.25×
32,768	1.00 GiB	1.25 GiB	1.25×
65,536	2.00 GiB	2.50 GiB	1.25×
131,072	4.00 GiB	5.00 GiB	1.25×