Is G9v3-3B or Qwen3-8B smaller?

G9v3-3B is smaller at Q4_K_M: 1.72 GiB against 4.68 GiB.

Which uses less memory at long context?

G9v3-3B — its KV cache at 32K context is 2.8x smaller, because of how its attention layers cache.

Model comparison

G9v3-3B vs Qwen3-8B

At Q4_K_M, G9v3-3B is the smaller download — 1,843,804,448 bytes against 5,027,783,488. At long context the gap widens: G9v3-3B's KV cache at 32K is 2.8× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	G9v3-3B	Qwen3-8B
Parameters	3.0B	8.2B
Architecture	llama	qwen3
Layers	52	36
Native context	131,072	40,960
Mixture of experts	no	no
Quantizations published	40	51
Smallest quantization	1.15 GiB	2.12 GiB
Q4_K_M	1.72 GiB	4.68 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	G9v3-3B	Qwen3-8B	Ratio
4,096	0.20 GiB	0.56 GiB	2.77×
8,192	0.41 GiB	1.13 GiB	2.77×
16,384	0.81 GiB	2.25 GiB	2.77×
32,768	1.63 GiB	4.50 GiB	2.77×
65,536	3.25 GiB	9.00 GiB	2.77×
131,072	6.50 GiB	18.00 GiB	2.77×