Is Qwen3.6-27B or phi-4 smaller?

phi-4 is smaller at Q4_K_M: 8.28 GiB against 15.41 GiB.

Which uses less memory at long context?

Qwen3.6-27B — its KV cache at 32K context is 3.1x smaller, because of how its attention layers cache.

Model comparison

Qwen3.6-27B vs phi-4

At Q4_K_M, phi-4 is the smaller download — 8,890,306,112 bytes against 16,547,397,888. At long context the gap widens: Qwen3.6-27B's KV cache at 32K is 3.1× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.6-27B	phi-4
Parameters	27.8B	14.7B
Architecture	qwen35	phi3
Layers	64	40
Native context	262,144	16,384
Mixture of experts	no	no
Quantizations published	40	37
Smallest quantization	8.74 GiB	4.18 GiB
Q4_K_M	15.41 GiB	8.28 GiB
Licence	apache-2.0	—

KV cache by context

the term that decides long-context viability

Context	Qwen3.6-27B	phi-4	Ratio
4,096	0.25 GiB	0.78 GiB	3.12×
8,192	0.50 GiB	1.56 GiB	3.12×
16,384	1.00 GiB	3.13 GiB	3.12×
32,768	2.00 GiB	6.25 GiB	3.12×
65,536	4.00 GiB	12.50 GiB	3.12×
131,072	8.00 GiB	25.00 GiB	3.12×