Is Qwen3-Coder-30B-A3B-Instruct or Qwen3.5-88B smaller?

Qwen3-Coder-30B-A3B-Instruct is smaller at Q4_K_M: 17.28 GiB against 49.66 GiB.

Which uses less memory at long context?

Qwen3.5-88B — its KV cache at 32K context is 4.0x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct vs Qwen3.5-88B

At Q4_K_M, Qwen3-Coder-30B-A3B-Instruct is the smaller download — 18,556,689,568 bytes against 53,323,531,328. At long context the gap widens: Qwen3.5-88B's KV cache at 32K is 4.0× 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	Qwen3.5-88B
Parameters	30.5B	87.7B
Architecture	qwen3moe	qwen35moe
Layers	48	48
Native context	262,144	262,144
Mixture of experts	yes, 128 experts	yes, 180 experts
Quantizations published	46	33
Smallest quantization	7.46 GiB	17.20 GiB
Q4_K_M	17.28 GiB	49.66 GiB
Licence	apache-2.0	other

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct	Qwen3.5-88B	Ratio
4,096	0.38 GiB	0.09 GiB	4.00×
8,192	0.75 GiB	0.19 GiB	4.00×
16,384	1.50 GiB	0.38 GiB	4.00×
32,768	3.00 GiB	0.75 GiB	4.00×
65,536	6.00 GiB	1.50 GiB	4.00×
131,072	12.00 GiB	3.00 GiB	4.00×