Is Qwen3.8-27B or reka-flash-3.1 smaller?

reka-flash-3.1 is smaller at Q4_K_M: 12.68 GiB against 15.66 GiB.

Which uses less memory at long context?

Qwen3.8-27B — its KV cache at 32K context is 2.1x smaller, because of how its attention layers cache.

Model comparison

Qwen3.8-27B vs reka-flash-3.1

At Q4_K_M, reka-flash-3.1 is the smaller download — 13,610,362,432 bytes against 16,810,714,336. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 2.1× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.8-27B	reka-flash-3.1
Parameters	27.8B	20.9B
Architecture	qwen35	llama
Layers	64	44
Native context	262,144	98,304
Mixture of experts	no	no
Quantizations published	22	51
Smallest quantization	8.39 GiB	6.15 GiB
Q4_K_M	15.66 GiB	12.68 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3.8-27B	reka-flash-3.1	Ratio
4,096	0.25 GiB	0.52 GiB	2.06×
8,192	0.50 GiB	1.03 GiB	2.06×
16,384	1.00 GiB	2.06 GiB	2.06×
32,768	2.00 GiB	4.13 GiB	2.06×
65,536	4.00 GiB	8.25 GiB	2.06×
131,072	8.00 GiB	16.50 GiB	2.06×