Model comparison

z_image_turbo vs gemma-3-270m-it

At Q4_K_M, gemma-3-270m-it is the smaller download — 253,115,168 bytes against 4,511,501,376.

From the file· summed bytes, KV per layer

Side by side

z_image_turbogemma-3-270m-it
Parameters84M268M
Architecturepiggemma3
Layers18
Native context32,768
Mixture of expertsnono
Quantizations published1739
Smallest quantization0.16 GiB0.17 GiB
Q4_K_M4.20 GiB0.24 GiB
Licenceapache-2.0gemma

KV cache by context

the term that decides long-context viability
Contextz_image_turbogemma-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