Model comparison

embeddinggemma-300m vs gemma-3-1b-it

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

embeddinggemma-300mgemma-3-1b-it
Parameters303M1000M
Architecturegemma-embeddinggemma3
Layers2426
Native context2,04832,768
Mixture of expertsnono
Quantizations published1028
Smallest quantization0.26 GiB0.52 GiB
Q4_K_M0.75 GiB
Licencegemma

KV cache by context

the term that decides long-context viability
Contextembeddinggemma-300mgemma-3-1b-itRatio
4,0960.04 GiB0.04 GiB1.06×
8,1920.05 GiB0.05 GiB1.04×
16,3840.08 GiB0.08 GiB1.02×
32,7680.14 GiB0.15 GiB1.01×
65,5360.27 GiB0.27 GiB1.01×
131,0720.52 GiB0.52 GiB1.00×