Is Qwen3.6-35B-A3B or Mistral-Small-3.2-24B-Instruct-2506 smaller?

Mistral-Small-3.2-24B-Instruct-2506 is smaller at Q4_K_M: 13.35 GiB against 19.71 GiB.

Which uses less memory at long context?

Qwen3.6-35B-A3B — its KV cache at 32K context is 8.0x smaller, because of how its attention layers cache.

Model comparison

Qwen3.6-35B-A3B vs Mistral-Small-3.2-24B-Instruct-2506

At Q4_K_M, Mistral-Small-3.2-24B-Instruct-2506 is the smaller download — 14,333,909,728 bytes against 21,166,757,728. At long context the gap widens: Qwen3.6-35B-A3B's KV cache at 32K is 8.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3.6-35B-A3B	Mistral-Small-3.2-24B-Instruct-2506
Parameters	36.0B	24.0B
Architecture	qwen35moe	llama
Layers	40	40
Native context	262,144	131,072
Mixture of experts	yes, 256 experts	no
Quantizations published	61	50
Smallest quantization	8.77 GiB	5.18 GiB
Q4_K_M	19.71 GiB	13.35 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3.6-35B-A3B	Mistral-Small-3.2-24B-Instruct-2506	Ratio
4,096	0.08 GiB	0.63 GiB	7.99×
8,192	0.16 GiB	1.25 GiB	8.00×
16,384	0.31 GiB	2.50 GiB	8.00×
32,768	0.63 GiB	5.00 GiB	8.00×
65,536	1.25 GiB	10.00 GiB	8.00×
131,072	2.50 GiB	20.00 GiB	8.00×