Is Qwen3.8-27B or GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill smaller?

Qwen3.8-27B is smaller at Q4_K_M: 15.66 GiB against 16.89 GiB.

Which uses less memory at long context?

GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill — its KV cache at 32K context is 1.2x smaller, because of how its attention layers cache.

Model comparison

Qwen3.8-27B vs GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill

At Q4_K_M, Qwen3.8-27B is the smaller download — 16,810,714,336 bytes against 18,132,721,856. At long context the gap widens: GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill'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

	Qwen3.8-27B	GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill
Parameters	27.8B	31.2B
Architecture	qwen35	deepseek2
Layers	64	47
Native context	262,144	202,752
Mixture of experts	no	yes, 64 experts
Quantizations published	22	13
Smallest quantization	8.39 GiB	9.22 GiB
Q4_K_M	15.66 GiB	16.89 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3.8-27B	GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill	Ratio
4,096	0.25 GiB	0.21 GiB	1.21×
8,192	0.50 GiB	0.41 GiB	1.21×
16,384	1.00 GiB	0.83 GiB	1.21×
32,768	2.00 GiB	1.65 GiB	1.21×
65,536	4.00 GiB	3.30 GiB	1.21×
131,072	8.00 GiB	6.61 GiB	1.21×