Model comparison

Ling-mini-2.0 vs gemma-4-E4B-it

At Q4_K_M, gemma-4-E4B-it is the smaller download — 4,977,171,584 bytes against 9,911,575,072. At long context the gap widens: gemma-4-E4B-it's KV cache at 32K is 2.5× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Ling-mini-2.0gemma-4-E4B-it
Parameters16.3B8.0B
Architecturebailingmoe2gemma4
Layers2042
Native context32,768131,072
Mixture of expertsyes, 256 expertsno
Quantizations published3036
Smallest quantization4.34 GiB3.30 GiB
Q4_K_M9.23 GiB4.64 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextLing-mini-2.0gemma-4-E4B-itRatio
4,0960.16 GiB0.12 GiB1.27×
8,1920.31 GiB0.18 GiB1.76×
16,3840.63 GiB0.29 GiB2.18×
32,7681.25 GiB0.51 GiB2.47×
65,5362.50 GiB0.94 GiB2.65×
131,0725.00 GiB1.82 GiB2.75×