Is GLM-4.6V-Flash or Qwen3-30B-A3B-Thinking-2507 smaller?

GLM-4.6V-Flash is smaller at Q4_K_M: 5.74 GiB against 17.28 GiB.

Which uses less memory at long context?

GLM-4.6V-Flash — its KV cache at 32K context is 2.4x smaller, because of how its attention layers cache.

Model comparison

GLM-4.6V-Flash vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, GLM-4.6V-Flash is the smaller download — 6,166,577,888 bytes against 18,556,685,824. At long context the gap widens: GLM-4.6V-Flash'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

	GLM-4.6V-Flash	Qwen3-30B-A3B-Thinking-2507
Parameters	10.3B	30.5B
Architecture	glm4	qwen3moe
Layers	40	48
Native context	131,072	262,144
Mixture of experts	no	yes, 128 experts
Quantizations published	47	51
Smallest quantization	2.98 GiB	7.05 GiB
Q4_K_M	5.74 GiB	17.28 GiB
Licence	mit	apache-2.0

KV cache by context

the term that decides long-context viability

Context	GLM-4.6V-Flash	Qwen3-30B-A3B-Thinking-2507	Ratio
4,096	0.16 GiB	0.38 GiB	2.40×
8,192	0.31 GiB	0.75 GiB	2.40×
16,384	0.63 GiB	1.50 GiB	2.40×
32,768	1.25 GiB	3.00 GiB	2.40×
65,536	2.50 GiB	6.00 GiB	2.40×
131,072	5.00 GiB	12.00 GiB	2.40×