Model comparison

Z-Image vs Qwen-Image-Edit-2511

At Q4_K_M, Z-Image is the smaller download — 5,066,995,776 bytes against 11,657,701,984.

From the file· summed bytes, KV per layer

Side by side

Z-ImageQwen-Image-Edit-2511
Parameters6.2B20.4B
Architecturelumina2qwen_image
Layers
Native context
Mixture of expertsnono
Quantizations published1636
Smallest quantization3.74 GiB4.13 GiB
Q4_K_M4.72 GiB10.86 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextZ-ImageQwen-Image-Edit-2511Ratio
4,096
8,192
16,384
32,768
65,536
131,072
Z-Image vs Qwen-Image-Edit-2511 — size, memory and hardware fit — ossmodeldb