Is Qwen3.8-27B or deepseek-llm-67b-base smaller?

Qwen3.8-27B is smaller at Q4_K_M: 15.66 GiB against 37.66 GiB.

Which uses less memory at long context?

Qwen3.8-27B — its KV cache at 32K context is 5.9x smaller, because of how its attention layers cache.

Model comparison

Qwen3.8-27B vs deepseek-llm-67b-base

At Q4_K_M, Qwen3.8-27B is the smaller download — 16,810,714,336 bytes against 40,439,686,592. At long context the gap widens: Qwen3.8-27B's KV cache at 32K is 5.9× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.8-27B	deepseek-llm-67b-base
Parameters	27.8B	67.4B
Architecture	qwen35	llama
Layers	64	95
Native context	262,144	4,096
Mixture of experts	no	no
Quantizations published	22	30
Smallest quantization	8.39 GiB	13.65 GiB
Q4_K_M	15.66 GiB	37.66 GiB
Licence	apache-2.0	other

KV cache by context

the term that decides long-context viability

Context	Qwen3.8-27B	deepseek-llm-67b-base	Ratio
4,096	0.25 GiB	1.48 GiB	5.93×
8,192	0.50 GiB	2.97 GiB	5.94×
16,384	1.00 GiB	5.94 GiB	5.94×
32,768	2.00 GiB	11.88 GiB	5.94×
65,536	4.00 GiB	23.75 GiB	5.94×
131,072	8.00 GiB	47.50 GiB	5.94×