Is Qwen3-Coder-30B-A3B-Instruct or Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored smaller?

Qwen3-Coder-30B-A3B-Instruct is smaller at Q4_K_M: 17.28 GiB against 22.83 GiB.

Which uses less memory at long context?

Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored — its KV cache at 32K context is 1.8x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct vs Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored

At Q4_K_M, Qwen3-Coder-30B-A3B-Instruct is the smaller download — 18,556,689,568 bytes against 24,515,131,200. At long context the gap widens: Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored's KV cache at 32K is 1.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Qwen3-Coder-30B-A3B-Instruct	Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored
Parameters	30.5B	31.6B
Architecture	qwen3moe	nemotron_h_moe
Layers	48	52
Native context	262,144	262,144
Mixture of experts	yes, 128 experts	yes, 128 experts
Quantizations published	46	34
Smallest quantization	7.46 GiB	16.64 GiB
Q4_K_M	17.28 GiB	22.83 GiB
Licence	apache-2.0	other

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct	Nemotron-Cascade-2-30B-A3B-heretic-ara-uncensored	Ratio
4,096	0.38 GiB	0.20 GiB	1.85×
8,192	0.75 GiB	0.41 GiB	1.85×
16,384	1.50 GiB	0.81 GiB	1.85×
32,768	3.00 GiB	1.63 GiB	1.85×
65,536	6.00 GiB	3.25 GiB	1.85×
131,072	12.00 GiB	6.50 GiB	1.85×