Shyvana VGU Rework 2026 | League of Legends
PICK THE RIGHT VERSION!
v1.0 - Illustrious
v2.0 - Pony
This LoRA was trained on the official in-game model of Shyvana’s 2026 Visual & Gameplay Update combined with high-quality shots from the "Forged in Fury" cinematic for accurate colors, lighting, and details.
Recommended Settings:
Lora Weight: 0.9 - 1.2
Sampler: Euler Ancestral
Scheduler: Normal or beta_1_1
Steps: 20-40
CFG Scale: 5-7
Resolution: 768x1024 (Start with this.)
Note: If you have having background problems, it dose not get generate, come our broken, etc... Switch to: DPMPP_2m and KARRAS
Trigger word: driftpixel_shyvana
Do not utilise the trigger word alone, the outcome will most likely come out half baked. You need to be very specific with your prompting in order to get a fantastic result.
Recommended Universal Positive Prompt:driftpixel_shyvana, dragon girl, purple skin, solo, 1girl, long hair, yellow eyes, blonde hair, red horns, orange sclera, yellow iris, reptilian pupils, pointy ears, abs, midriff, armor, silver breastplate, brown leather pants, spike on shoulders
For Pony just add this in between the trigger word and your prompt: score_9, score_8_up, score_7_up, score_6_up
Recommended Universal Negative Prompt:
worst quality, low quality, normal quality, lowres, blurry, jpeg artifacts, compression artifacts, bad anatomy, bad hands, extra limbs, fused fingers, missing fingers, too many fingers, deformed hands, mutated hands, poorly drawn hands, extra arms, extra legs, fused limbs, mutated limbs, disconnected limbs, bad proportions, gross proportions, cloned face, duplicate, two girls with same face, identical twins, same character, duplicate character, same hair, same eyes, same body, brown skin, tan skin, orange skin, yellow skin, washed out colors, desaturated, overexposed, underexposed, bad lighting, flat lighting, watermark, text, signature, logo, username, artist name, censored, bar censor, mosaic censor, deformed strap-on, broken strap-on, floating strap-on, extra dildo, bad penetration, impossible anatomy, floating limbs, error, cropped, out of frame
For Pony add something like this ad the beginning of your prompt: score_5, score_4, score_3,
ComfyUI Workflow Included!
Just download an sample image i uploaded and simply DRAG & DROP it into the ComfyUI interface... the images i uploaded should still hold the workflow metadata.
But if it's not, here's is the link to my workflow post: CLICK HERE
This represents many hours of work. If you enjoy it, please 👍like, 💬 comment , and feel free to ⚡tip!
Description
Base model: Illustrious Strange seduction
BAD VERSION, DO NOT USE!
This is my first LoRA upload publicly and i overcooked it a little bit, it's hard to work with it overall. Poses are inconsistent, hard to change the style or try a dynamic pose.
Details
Files
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.


