Is Hemlock-Qwopus3.5-9B-Coder or gemma-4-26B-A4B-it smaller?

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

Which uses less memory at long context?

Hemlock-Qwopus3.5-9B-Coder — its KV cache at 32K context is 1.5x smaller, because of how its attention layers cache.

Model comparison

Hemlock-Qwopus3.5-9B-Coder vs gemma-4-26B-A4B-it

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: Hemlock-Qwopus3.5-9B-Coder's KV cache at 32K is 1.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	Hemlock-Qwopus3.5-9B-Coder	gemma-4-26B-A4B-it
Parameters	9.7B	26.5B
Architecture	qwen35	gemma4
Layers	32	30
Native context	262,144	262,144
Mixture of experts	no	yes, 128 experts
Quantizations published	15	46
Smallest quantization	3.65 GiB	8.99 GiB
Q4_K_M	—	15.64 GiB
Licence	—	apache-2.0

KV cache by context

the term that decides long-context viability

Context	Hemlock-Qwopus3.5-9B-Coder	gemma-4-26B-A4B-it	Ratio
4,096	0.13 GiB	0.45 GiB	3.59×
8,192	0.25 GiB	0.61 GiB	2.42×
16,384	0.50 GiB	0.92 GiB	1.84×
32,768	1.00 GiB	1.54 GiB	1.54×
65,536	2.00 GiB	2.79 GiB	1.40×
131,072	4.00 GiB	5.29 GiB	1.32×