Is EXAONE-4.5-33B or Qwen3.6-35B-A3B smaller?

EXAONE-4.5-33B is smaller at Q4_K_M: 18.67 GiB against 19.71 GiB.

Which uses less memory at long context?

Qwen3.6-35B-A3B — its KV cache at 32K context is 4.5x smaller, because of how its attention layers cache.

Model comparison

EXAONE-4.5-33B vs Qwen3.6-35B-A3B

At Q4_K_M, EXAONE-4.5-33B is the smaller download — 20,047,839,424 bytes against 21,166,757,728. At long context the gap widens: Qwen3.6-35B-A3B's KV cache at 32K is 4.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	EXAONE-4.5-33B	Qwen3.6-35B-A3B
Parameters	34.4B	36.0B
Architecture	exaone4	qwen35moe
Layers	64	40
Native context	262,144	262,144
Mixture of experts	no	yes, 256 experts
Quantizations published	31	61
Smallest quantization	11.57 GiB	8.77 GiB
Q4_K_M	18.67 GiB	19.71 GiB
Licence	other	apache-2.0

KV cache by context

the term that decides long-context viability

Context	EXAONE-4.5-33B	Qwen3.6-35B-A3B	Ratio
4,096	1.00 GiB	0.08 GiB	12.79×
8,192	1.34 GiB	0.16 GiB	8.60×
16,384	1.84 GiB	0.31 GiB	5.90×
32,768	2.84 GiB	0.63 GiB	4.55×
65,536	4.84 GiB	1.25 GiB	3.87×
131,072	8.84 GiB	2.50 GiB	3.54×