Is internlm2-math-plus-20b or Qwen3.8-27B smaller?

internlm2-math-plus-20b is smaller at Q4_K_M: 11.16 GiB against 15.66 GiB.

Which uses less memory at long context?

Qwen3.8-27B — its KV cache at 32K context is 3.0x smaller, because of how its attention layers cache.

Model comparison

internlm2-math-plus-20b vs Qwen3.8-27B

At Q4_K_M, internlm2-math-plus-20b is the smaller download — 11,984,469,440 bytes against 16,810,714,336. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 3.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	internlm2-math-plus-20b	Qwen3.8-27B
Parameters	19.9B	27.8B
Architecture	internlm2	qwen35
Layers	48	64
Native context	8,192	262,144
Mixture of experts	no	no
Quantizations published	68	22
Smallest quantization	4.23 GiB	8.39 GiB
Q4_K_M	11.16 GiB	15.66 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	internlm2-math-plus-20b	Qwen3.8-27B	Ratio
4,096	0.75 GiB	0.25 GiB	3.00×
8,192	1.50 GiB	0.50 GiB	3.00×
16,384	3.00 GiB	1.00 GiB	3.00×
32,768	6.00 GiB	2.00 GiB	3.00×
65,536	12.00 GiB	4.00 GiB	3.00×
131,072	24.00 GiB	8.00 GiB	3.00×