Model comparison

next-1b vs gemma-3-1b-it

At Q4_K_M, next-1b is the smaller download — 806,057,664 bytes against 806,058,240.

From the file· summed bytes, KV per layer

Side by side

next-1bgemma-3-1b-it
Parameters1000M1000M
Architecturegemma3gemma3
Layers2626
Native context32,76832,768
Mixture of expertsnono
Quantizations published128
Smallest quantization0.75 GiB0.52 GiB
Q4_K_M0.75 GiB0.75 GiB
Licencemitgemma

KV cache by context

the term that decides long-context viability
Contextnext-1bgemma-3-1b-itRatio
4,0960.04 GiB0.04 GiB1.00×
8,1920.05 GiB0.05 GiB1.00×
16,3840.08 GiB0.08 GiB1.00×
32,7680.15 GiB0.15 GiB1.00×
65,5360.27 GiB0.27 GiB1.00×
131,0720.52 GiB0.52 GiB1.00×