Model comparison

Janus-Pro-7B vs Z-Image-Turbo

At Q4_K_M, Janus-Pro-7B is the smaller download — 4,223,359,936 bytes against 4,981,532,736.

From the file· summed bytes, KV per layer

Side by side

Janus-Pro-7BZ-Image-Turbo
Parameters7.4B6.2B
Architecturellamalumina2
Layers30
Native context16,384
Mixture of expertsnono
Quantizations published3630
Smallest quantization1.61 GiB2.41 GiB
Q4_K_M3.93 GiB4.64 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextJanus-Pro-7BZ-Image-TurboRatio
4,0961.88 GiB
8,1923.75 GiB
16,3847.50 GiB
32,76815.00 GiB
65,53630.00 GiB
131,07260.00 GiB
Janus-Pro-7B vs Z-Image-Turbo — size, memory and hardware fit — ossmodeldb