Is EXAONE-4.0-1.2B-abliterated 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 13.0x smaller, because of how its attention layers cache.

Model comparison

EXAONE-4.0-1.2B-abliterated 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 13.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	EXAONE-4.0-1.2B-abliterated	embeddinggemma-300m
Parameters	1.5B	303M
Architecture	exaone4	gemma-embedding
Layers	30	24
Native context	65,536	2,048
Mixture of experts	no	no
Quantizations published	36	10
Smallest quantization	0.42 GiB	0.26 GiB
Q4_K_M	0.87 GiB	—
Licence	—	—

KV cache by context

the term that decides long-context viability

Context	EXAONE-4.0-1.2B-abliterated	embeddinggemma-300m	Ratio
4,096	0.23 GiB	0.04 GiB	6.67×
8,192	0.47 GiB	0.05 GiB	9.23×
16,384	0.94 GiB	0.08 GiB	11.43×
32,768	1.88 GiB	0.14 GiB	12.97×
65,536	3.75 GiB	0.27 GiB	13.91×
131,072	7.50 GiB	0.52 GiB	14.44×