Is GLM-4.7-Flash-Derestricted or Qwen3-30B-A3B-Thinking-2507 smaller?

GLM-4.7-Flash-Derestricted is smaller at Q4_K_M: 16.89 GiB against 17.28 GiB.

Which uses less memory at long context?

GLM-4.7-Flash-Derestricted — its KV cache at 32K context is 1.8x smaller, because of how its attention layers cache.

Model comparison

GLM-4.7-Flash-Derestricted vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, GLM-4.7-Flash-Derestricted is the smaller download — 18,132,722,080 bytes against 18,556,685,824. At long context the gap widens: GLM-4.7-Flash-Derestricted's KV cache at 32K is 1.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	GLM-4.7-Flash-Derestricted	Qwen3-30B-A3B-Thinking-2507
Parameters	31.2B	30.5B
Architecture	deepseek2	qwen3moe
Layers	47	48
Native context	202,752	262,144
Mixture of experts	yes, 64 experts	yes, 128 experts
Quantizations published	34	51
Smallest quantization	5.78 GiB	7.05 GiB
Q4_K_M	16.89 GiB	17.28 GiB
Licence	mit	apache-2.0

KV cache by context

the term that decides long-context viability

Context	GLM-4.7-Flash-Derestricted	Qwen3-30B-A3B-Thinking-2507	Ratio
4,096	0.21 GiB	0.38 GiB	1.82×
8,192	0.41 GiB	0.75 GiB	1.82×
16,384	0.83 GiB	1.50 GiB	1.82×
32,768	1.65 GiB	3.00 GiB	1.82×
65,536	3.30 GiB	6.00 GiB	1.82×
131,072	6.61 GiB	12.00 GiB	1.82×