Model comparison

ced-base vs gemma-3-270m-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

ced-basegemma-3-270m-it
Parameters86M268M
Architecturecedgemma3
Layers18
Native context32,768
Mixture of expertsnono
Quantizations published339
Smallest quantization0.12 GiB0.17 GiB
Q4_K_M0.24 GiB
Licenceapache-2.0gemma

KV cache by context

the term that decides long-context viability
Contextced-basegemma-3-270m-itRatio
4,0960.03 GiB
8,1920.04 GiB
16,3840.06 GiB
32,7680.11 GiB
65,5360.20 GiB
131,0720.39 GiB