Model comparison

stable-diffusion-3.5-medium vs FLUX.2-klein-4B

At Q4_K_M, stable-diffusion-3.5-medium is the smaller download — 1,787,064,768 bytes against 2,497,280,416.

From the file· summed bytes, KV per layer

Side by side

stable-diffusion-3.5-mediumFLUX.2-klein-4B
Parameters2.5B3.9B
Architectureflux
Layers
Native context
Mixture of expertsnono
Quantizations published2121
Smallest quantization1.35 GiB1.70 GiB
Q4_K_M1.66 GiB2.33 GiB
Licenceotherapache-2.0

KV cache by context

the term that decides long-context viability
Contextstable-diffusion-3.5-mediumFLUX.2-klein-4BRatio
4,096
8,192
16,384
32,768
65,536
131,072