Short description

A character LoRA for Dizzy, intended to preserve the character's recognizable identity across different Illustrious checkpoints and scenes. Use it as the identity layer, then let the base checkpoint or an additional low-weight style LoRA control the final visual finish.

Generation parameters

• Base model(s): Illustrious
• Trigger(s): Illustrious V1: dizzy_(character)
• LoRA weight: 1.0 (recommended starting point; lower it mainly when stacking multiple LoRAs or when you want a softer style influence)
• Sampler: DPM++ 2M SDE Karras; try Euler a when you want a rougher or more graphic interpretation
• Steps: 28–36
• CFG: 4.5–6
• Resolution: 1024×1024, 832×1216, or 1216×832
• If the style begins to overpower character identity or composition, reduce LoRA weight before changing CFG.

Prompt examples

TODO: Add 2–3 prompt examples validated against the current version.

Give me a hand

If you use Dizzy across different checkpoints, outfits, poses, or scenes, please post the generations. Seeing where the character identity stays stable—and where it starts drifting—is exactly the feedback that helps improve a character LoRA; an upvote is appreciated if it proves useful.

Suggestions

Recommended base from this profile: SweetIllustrious. For a controlled style blend, try ItsUnkownAnon style | [Illustrious] at a low secondary weight (0.2–0.45) and keep this LoRA as the dominant style.