Is gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant or embeddinggemma-300m smaller?

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

Which uses less memory at long context?

embeddinggemma-300m — its KV cache at 32K context is 2.0x smaller, because of how its attention layers cache.

Model comparison

gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant vs embeddinggemma-300m

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: embeddinggemma-300m'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-4-26B-A4B-it-qat-q4_0-unquantized-assistant	embeddinggemma-300m
Parameters	420M	303M
Architecture	gemma4-assistant	gemma-embedding
Layers	4	24
Native context	262,144	2,048
Mixture of experts	no	no
Quantizations published	11	10
Smallest quantization	0.27 GiB	0.26 GiB
Q4_K_M	0.31 GiB	—
Licence	mit	—

KV cache by context

the term that decides long-context viability

Context	gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant	embeddinggemma-300m	Ratio
4,096	0.07 GiB	0.04 GiB	1.89×
8,192	0.10 GiB	0.05 GiB	1.92×
16,384	0.16 GiB	0.08 GiB	1.95×
32,768	0.29 GiB	0.14 GiB	1.97×
65,536	0.54 GiB	0.27 GiB	1.99×
131,072	1.04 GiB	0.52 GiB	1.99×