Is Qwen3.6-27B or Qwen3-Coder-REAP-25B-A3B smaller?

Qwen3-Coder-REAP-25B-A3B is smaller at Q4_K_M: 14.08 GiB against 15.41 GiB.

Which uses less memory at long context?

Qwen3.6-27B — its KV cache at 32K context is 1.5x smaller, because of how its attention layers cache.

Model comparison

Qwen3.6-27B vs Qwen3-Coder-REAP-25B-A3B

At Q4_K_M, Qwen3-Coder-REAP-25B-A3B is the smaller download — 15,118,506,592 bytes against 16,547,397,888. At long context the gap widens: Qwen3.6-27B's KV cache at 32K is 1.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.6-27B	Qwen3-Coder-REAP-25B-A3B
Parameters	27.8B	24.9B
Architecture	qwen35	qwen3moe
Layers	64	48
Native context	262,144	262,144
Mixture of experts	no	yes, 103 experts
Quantizations published	40	62
Smallest quantization	8.74 GiB	4.91 GiB
Q4_K_M	15.41 GiB	14.08 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3.6-27B	Qwen3-Coder-REAP-25B-A3B	Ratio
4,096	0.25 GiB	0.38 GiB	1.50×
8,192	0.50 GiB	0.75 GiB	1.50×
16,384	1.00 GiB	1.50 GiB	1.50×
32,768	2.00 GiB	3.00 GiB	1.50×
65,536	4.00 GiB	6.00 GiB	1.50×
131,072	8.00 GiB	12.00 GiB	1.50×