Sign In

Verified:

SafeTensor

Type

LoRA

Stats

43

0

Reviews

Published

Jan 4, 2025

Base Model

Other

Training

Steps: 65,682
Epochs: 45

Usage Tips

Clip Skip: 1
Strength: 1

Trigger Words

NONE

Hash

AutoV2
344A436560

LoRA Description:

This LoRA is designed to convert realistic images into anime-style visuals. The key trigger words are source_anime, source_realistic, and BeyondNegative.

Known Issues:

  1. When using the trigger word source_anime, the LoRA currently produces a single style and character type. This limitation is being addressed and will be resolved in the next update with support for multiple styles (e.g., anime_style1, anime_style2, etc.).

  2. The BeyondNegative trigger word can conflict with some other trigger words. This issue will also be addressed in the next update.

Compatibility:

  • This is a beta LoRA specifically developed to work with the Beyond Experimental models (latest version).

  • Please note: This LoRA is not compatible with pony models.

Additional Trigger Words: While the main trigger words are listed above, additional trigger words are available. Click the Generate button to explore the full range of options.