Model comparison

gliner2-base-v1 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

gliner2-base-v1gemma-3-270m-it
Parameters208M268M
Architecturedebertagemma3
Layers18
Native context32,768
Mixture of expertsnono
Quantizations published139
Smallest quantization0.13 GiB0.17 GiB
Q4_K_M0.24 GiB
Licencegemma

KV cache by context

the term that decides long-context viability
Contextgliner2-base-v1gemma-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