Model comparison

FLUX.2-klein-4B vs Chroma

At Q4_K_M, FLUX.2-klein-4B is the smaller download — 2,497,280,416 bytes against 5,566,533,792.

From the file· summed bytes, KV per layer

Side by side

FLUX.2-klein-4BChroma
Parameters3.9B8.9B
Architecturefluxflux
Layers
Native context
Mixture of expertsnono
Quantizations published21273
Smallest quantization1.70 GiB3.99 GiB
Q4_K_M2.33 GiB5.18 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextFLUX.2-klein-4BChromaRatio
4,096
8,192
16,384
32,768
65,536
131,072
FLUX.2-klein-4B vs Chroma — size, memory and hardware fit — ossmodeldb