Model comparison

ERNIE-Image vs FLUX.1-dev

At Q4_K_M, ERNIE-Image is the smaller download — 2,146,489,664 bytes against 6,931,817,760.

From the file· summed bytes, KV per layer

Side by side

ERNIE-ImageFLUX.1-dev
Parameters8.0B11.9B
Architecturewanflux
Layers
Native context
Mixture of expertsnono
Quantizations published2937
Smallest quantization2.00 GiB3.75 GiB
Q4_K_M2.00 GiB6.46 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
ContextERNIE-ImageFLUX.1-devRatio
4,096
8,192
16,384
32,768
65,536
131,072