Is functiongemma-270m-it or SmolLM2-135M-Instruct smaller?

SmolLM2-135M-Instruct is smaller at Q4_K_M: 0.10 GiB against 0.24 GiB.

Which uses less memory at long context?

functiongemma-270m-it — its KV cache at 32K context is 6.5x smaller, because of how its attention layers cache.

Model comparison

functiongemma-270m-it vs SmolLM2-135M-Instruct

At Q4_K_M, SmolLM2-135M-Instruct is the smaller download — 105,454,144 bytes against 253,127,392. At long context the gap widens: functiongemma-270m-it's KV cache at 32K is 6.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	functiongemma-270m-it	SmolLM2-135M-Instruct
Parameters	268M	135M
Architecture	gemma3	llama
Layers	18	30
Native context	32,768	8,192
Mixture of experts	no	no
Quantizations published	45	41
Smallest quantization	0.17 GiB	0.08 GiB
Q4_K_M	0.24 GiB	0.10 GiB
Licence	gemma	apache-2.0

KV cache by context

the term that decides long-context viability

Context	functiongemma-270m-it	SmolLM2-135M-Instruct	Ratio
4,096	0.03 GiB	0.09 GiB	3.33×
8,192	0.04 GiB	0.18 GiB	4.62×
16,384	0.06 GiB	0.35 GiB	5.71×
32,768	0.11 GiB	0.70 GiB	6.49×
65,536	0.20 GiB	1.41 GiB	6.96×
131,072	0.39 GiB	2.81 GiB	7.22×