Type: LORA

Model introduction: LORA adopts a high-tech store, high-definition pictures are selected (2K-4K), and the training pictures are derived from high-definition photography. Very fine and reasonable structure are the selection conditions. This LORA is mainly controlled on elements.

Base film recommendations: (large realistic models)

Model refining base film: Laowang_Architecutral_MIX V0.5

Recommended: Laowang_Architecutral_MIX V0.5

https://www.liblib.ai/modelinfo/f111a6dc2fa74e07abe4fc2f73501f0d

Sampling Steps : 40-50

Sampling method : Euler

Size recommendation: 512*512丨768*512丨512*768 for both horizontal and vertical orientations (high-resolution restoration can be enabled)

Amplification algorithm:

8x-NMKD-Superscale

Basic Algorithm

Basic Algorithm 1.5

Trigger Words

Modern