Model comparison

Janus-Pro-7B vs Chroma

At Q4_K_M, Janus-Pro-7B is the smaller download — 4,223,359,936 bytes against 5,566,533,792.

From the file· summed bytes, KV per layer

Side by side

Janus-Pro-7BChroma
Parameters7.4B8.9B
Architecturellamaflux
Layers30
Native context16,384
Mixture of expertsnono
Quantizations published36273
Smallest quantization1.61 GiB3.99 GiB
Q4_K_M3.93 GiB5.18 GiB
Licencemitapache-2.0

KV cache by context

the term that decides long-context viability
ContextJanus-Pro-7BChromaRatio
4,0961.88 GiB
8,1923.75 GiB
16,3847.50 GiB
32,76815.00 GiB
65,53630.00 GiB
131,07260.00 GiB