Is gemma-3-270m-it-qat or embeddinggemma-300m smaller?

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

Which uses less memory at long context?

gemma-3-270m-it-qat — its KV cache at 32K context is 1.3x smaller, because of how its attention layers cache.

Model comparison

gemma-3-270m-it-qat vs embeddinggemma-300m

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: gemma-3-270m-it-qat's KV cache at 32K is 1.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	gemma-3-270m-it-qat	embeddinggemma-300m
Parameters	268M	303M
Architecture	gemma3	gemma-embedding
Layers	18	24
Native context	32,768	2,048
Mixture of experts	no	no
Quantizations published	1	10
Smallest quantization	0.22 GiB	0.26 GiB
Q4_K_M	—	—
Licence	—	—

KV cache by context

the term that decides long-context viability

Context	gemma-3-270m-it-qat	embeddinggemma-300m	Ratio
4,096	0.03 GiB	0.04 GiB	1.33×
8,192	0.04 GiB	0.05 GiB	1.33×
16,384	0.06 GiB	0.08 GiB	1.33×
32,768	0.11 GiB	0.14 GiB	1.33×
65,536	0.20 GiB	0.27 GiB	1.33×
131,072	0.39 GiB	0.52 GiB	1.33×