Is Qwen3.8-27B or HarmonicHarlequin_v5-20B smaller?

Qwen3.8-27B is smaller at Q4_K_M: 15.66 GiB against 18.71 GiB.

Which uses less memory at long context?

Qwen3.8-27B — its KV cache at 32K context is 32.5x smaller, because of how its attention layers cache.

Model comparison

Qwen3.8-27B vs HarmonicHarlequin_v5-20B

At Q4_K_M, Qwen3.8-27B is the smaller download — 16,810,714,336 bytes against 20,087,800,544. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 32.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.8-27B	HarmonicHarlequin_v5-20B
Parameters	27.8B	33.3B
Architecture	qwen35	llama
Layers	64	104
Native context	262,144	4,096
Mixture of experts	no	no
Quantizations published	22	35
Smallest quantization	8.39 GiB	6.77 GiB
Q4_K_M	15.66 GiB	18.71 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	Qwen3.8-27B	HarmonicHarlequin_v5-20B	Ratio
4,096	0.25 GiB	8.13 GiB	32.48×
8,192	0.50 GiB	16.25 GiB	32.49×
16,384	1.00 GiB	32.50 GiB	32.49×
32,768	2.00 GiB	65.00 GiB	32.50×
65,536	4.00 GiB	130.00 GiB	32.50×
131,072	8.00 GiB	260.00 GiB	32.50×