Is Darwin-4B-Chimera or Qwen3-8B smaller?

Darwin-4B-Chimera is smaller at Q4_K_M: 2.33 GiB against 4.68 GiB.

Which uses less memory at long context?

Darwin-4B-Chimera — its KV cache at 32K context is 3.5x smaller, because of how its attention layers cache.

Model comparison

Darwin-4B-Chimera vs Qwen3-8B

At Q4_K_M, Darwin-4B-Chimera is the smaller download — 2,497,281,088 bytes against 5,027,783,488. At long context the gap widens: Darwin-4B-Chimera's KV cache at 32K is 3.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Darwin-4B-Chimera	Qwen3-8B
Parameters	4.0B	8.2B
Architecture	qwen3	qwen3
Layers	36	36
Native context	40,960	40,960
Mixture of experts	no	no
Quantizations published	59	51
Smallest quantization	0.98 GiB	2.12 GiB
Q4_K_M	2.33 GiB	4.68 GiB
Licence	gemma	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Darwin-4B-Chimera	Qwen3-8B	Ratio
4,096	0.56 GiB	0.56 GiB	1.00×
8,192	0.71 GiB	1.13 GiB	1.57×
16,384	0.90 GiB	2.25 GiB	2.49×
32,768	1.28 GiB	4.50 GiB	3.52×
65,536	2.03 GiB	9.00 GiB	4.44×
131,072	3.53 GiB	18.00 GiB	5.10×