LoRA trained on Inputwo's <a target="_blank" rel="ugc" href="https://x.com/inputwo">(@inputwo)</a> art. All preview images use Hires.fix Upscale Factor: 2 Hires steps: 10 Denoising Strength: 0.2 Upscaler: 4x-AnimeSharp Hires.fix guide and upscaler here: <a rel="ugc" href="https://civarchive.com/models/78396/fix-your-blur-hires-models">Fix your Blur - HiRes Models</a> Use with <a target="_blank" rel="ugc" href="https://civarchive.com/models/288584?modelVersionId=324619">AutismMix_pony</a> for best results.

Inputwo style PonyXL

I actually took the time to learn more about training LoRAs. This version adheres to Inputwo's style far more than the previous version. This LoRA creates more visually pleasing images with <a rel="ugc" href="https://civitai.com/models/288584?modelVersionId=324619">AutismMix_pony</a> while <a rel="ugc" href="https://civitai.com/models/257749?modelVersionId=290640">Pony Diffusion XL v6</a> produces images that resemble Inputwo's style more accurately

Files with SHA256: fd7585731cc8230a8db26d2b8da2f4a7682b0dba30952d9dfa07236b63c21ee5