Is lift or gemma-4-26B-A4B-it smaller?

lift is smaller at Q4_K_M: 5.38 GiB against 15.64 GiB.

Which uses less memory at long context?

lift — its KV cache at 32K context is 1.5x smaller, because of how its attention layers cache.

Model comparison

lift vs gemma-4-26B-A4B-it

At Q4_K_M, lift is the smaller download — 5,780,090,336 bytes against 16,796,017,248. At long context the gap widens: lift'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

	lift	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	17	46
Smallest quantization	3.65 GiB	8.99 GiB
Q4_K_M	5.38 GiB	15.64 GiB
Licence	openrail	apache-2.0

KV cache by context

the term that decides long-context viability

Context	lift	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×