Is Qwen3.8-27B or GLM-Z1-Rumination-32B-0414 smaller?

Qwen3.8-27B is smaller at Q4_K_M: 15.66 GiB against 18.66 GiB.

Which uses less memory at long context?

Qwen3.8-27B — its KV cache at 32K context is 3.8x smaller, because of how its attention layers cache.

Model comparison

Qwen3.8-27B vs GLM-Z1-Rumination-32B-0414

At Q4_K_M, Qwen3.8-27B is the smaller download — 16,810,714,336 bytes against 20,040,332,576. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 3.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.8-27B	GLM-Z1-Rumination-32B-0414
Parameters	27.8B	33.1B
Architecture	qwen35	glm4
Layers	64	61
Native context	262,144	131,072
Mixture of experts	no	no
Quantizations published	22	29
Smallest quantization	8.39 GiB	9.45 GiB
Q4_K_M	15.66 GiB	18.66 GiB
Licence	apache-2.0	mit

KV cache by context

the term that decides long-context viability

Context	Qwen3.8-27B	GLM-Z1-Rumination-32B-0414	Ratio
4,096	0.25 GiB	0.95 GiB	3.81×
8,192	0.50 GiB	1.91 GiB	3.81×
16,384	1.00 GiB	3.81 GiB	3.81×
32,768	2.00 GiB	7.63 GiB	3.81×
65,536	4.00 GiB	15.25 GiB	3.81×
131,072	8.00 GiB	30.50 GiB	3.81×