Is gemma-2-9b-it-bnb-4bit or Qwen3-30B-A3B-Thinking-2507 smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

Qwen3-30B-A3B-Thinking-2507 — its KV cache at 32K context is 2.0x smaller, because of how its attention layers cache.

Model comparison

gemma-2-9b-it-bnb-4bit vs Qwen3-30B-A3B-Thinking-2507

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: Qwen3-30B-A3B-Thinking-2507's KV cache at 32K is 2.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-2-9b-it-bnb-4bit	Qwen3-30B-A3B-Thinking-2507
Parameters	9.5B	30.5B
Architecture	gemma2	qwen3moe
Layers	42	48
Native context	8,192	262,144
Mixture of experts	no	yes, 128 experts
Quantizations published	1	51
Smallest quantization	9.15 GiB	7.05 GiB
Q4_K_M	—	17.28 GiB
Licence	apache-2.0	apache-2.0

KV cache by context

the term that decides long-context viability

Context	gemma-2-9b-it-bnb-4bit	Qwen3-30B-A3B-Thinking-2507	Ratio
4,096	1.31 GiB	0.38 GiB	3.50×
8,192	2.05 GiB	0.75 GiB	2.73×
16,384	3.36 GiB	1.50 GiB	2.24×
32,768	5.99 GiB	3.00 GiB	2.00×
65,536	11.24 GiB	6.00 GiB	1.87×
131,072	21.74 GiB	12.00 GiB	1.81×