Is Qwen3-Coder-30B-A3B-Instruct or v6-Finch-14B-HF smaller?

v6-Finch-14B-HF is smaller at Q4_K_M: 8.17 GiB against 17.28 GiB.

Which uses less memory at long context?

Qwen3-Coder-30B-A3B-Instruct — its KV cache at 32K context is 10.2x smaller, because of how its attention layers cache.

Model comparison

Qwen3-Coder-30B-A3B-Instruct vs v6-Finch-14B-HF

At Q4_K_M, v6-Finch-14B-HF is the smaller download — 8,767,898,176 bytes against 18,556,689,568. At long context the gap widens: Qwen3-Coder-30B-A3B-Instruct's KV cache at 32K is 10.2× 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	v6-Finch-14B-HF
Parameters	30.5B	14.1B
Architecture	qwen3moe	rwkv6
Layers	48	61
Native context	262,144	—
Mixture of experts	yes, 128 experts	no
Quantizations published	46	20
Smallest quantization	7.46 GiB	5.10 GiB
Q4_K_M	17.28 GiB	8.17 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Qwen3-Coder-30B-A3B-Instruct	v6-Finch-14B-HF	Ratio
4,096	0.38 GiB	3.81 GiB	10.17×
8,192	0.75 GiB	7.63 GiB	10.17×
16,384	1.50 GiB	15.25 GiB	10.17×
32,768	3.00 GiB	30.50 GiB	10.17×
65,536	6.00 GiB	61.00 GiB	10.17×
131,072	12.00 GiB	122.00 GiB	10.17×