Is ALIA-40b-fc-2606 or Qwen3-30B-A3B-Thinking-2507 smaller?

Qwen3-30B-A3B-Thinking-2507 is smaller at Q4_K_M: 17.28 GiB against 22.90 GiB.

Which uses less memory at long context?

Qwen3-30B-A3B-Thinking-2507 — its KV cache at 32K context is 2.0x smaller, because of how its attention layers cache.

Model comparison

ALIA-40b-fc-2606 vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, Qwen3-30B-A3B-Thinking-2507 is the smaller download — 18,556,685,824 bytes against 24,591,543,264. At long context the gap widens: Qwen3-30B-A3B-Thinking-2507's KV cache at 32K is 2.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	ALIA-40b-fc-2606	Qwen3-30B-A3B-Thinking-2507
Parameters	40.4B	30.5B
Architecture	llama	qwen3moe
Layers	48	48
Native context	163,840	262,144
Mixture of experts	no	yes, 128 experts
Quantizations published	34	51
Smallest quantization	9.06 GiB	7.05 GiB
Q4_K_M	22.90 GiB	17.28 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	ALIA-40b-fc-2606	Qwen3-30B-A3B-Thinking-2507	Ratio
4,096	0.75 GiB	0.38 GiB	2.00×
8,192	1.50 GiB	0.75 GiB	2.00×
16,384	3.00 GiB	1.50 GiB	2.00×
32,768	6.00 GiB	3.00 GiB	2.00×
65,536	12.00 GiB	6.00 GiB	2.00×
131,072	24.00 GiB	12.00 GiB	2.00×