Sign In
homemodelsimagesvideos3D Modelshubsarticlescomicscruciblesupdatesshop

avaka-viveka-dancer SDXL - V1 Showcase

vivekchahaman600's Avatar

vivekchahaman600

4QXEFV96FPWSANPVPYQVG3H590.jpeg
XPKRVD7BWTF7ZMWYK9RN1QQ5T0.jpeg
Z2H69SEMFS90258SBENPTDQ780.jpeg
pose_18.jpg
pose_15.jpg

avaka-viveka-dancer

Character LoRA for a classical Indian temple dancer. Use the trigger word in every prompt. Keep the rest of the prompt short.

1. Quick start

  • Load your base checkpoint first (SDXL any version)

  • Add the LoRA. Start at weight 0.7–0.85. Raise only if the face or costume fades.

  • Put the trigger first in the positive prompt.

  • Use a full-body or mid-shot framing so limbs stay readable.

  • Keep negatives lean. Extra limbs and blur are the main failure modes.

2. Trigger and weights

Item

Value

Trigger word

avaka-viveka-dancer

Recommended weight

0.70 – 0.85 (try 1.0 only on a weak base)

Clip skip

1 for SDXL/Flux · 2 if the base is an anime/Illustrious model

Resolution

1024×1024 or 768×1280 portrait. Avoid tiny crops.

3. Prompt recipe

Positive (keep it short):

avaka-viveka-dancer, sitting in temple

Negative:

blurry, multi-limbed, extra arms, extra legs, extra fingers, deformed hands, cropped head, lowres, watermark

Formula: trigger + one pose + one place. Do not stack five style tags on the first try.

Comments