<h2 id="vengeful-spirits-and-dark-arts-by-hailoknight-5mba1ewg6">Vengeful Spirits &amp; Dark Arts - By HailoKnight👻</h2><h3 id="hello-friends!-ay9bojdyn">Hello friends!</h3><h3 id="this-lora-is-designed-to-give-your-images-that-extra-creepy-and-dark-vibes.-4w6x7a19p">This LoRA is designed to give your images that extra creepy and dark vibes.</h3><h3 id="things-to-know:-65oc30r2w">Things To know:</h3><ul><li>You should use the trigger word hkhorror</li><li>MOST IMAGE ARE DONE WITH <a target="_blank" rel="ugc" href="https://civarchive.com/models/143043/starlight-xl-animated">STARLIGHT</a><a target="_blank" rel="ugc" href="https://civarchive.com/models/391165/ponyplex?modelVersionId=436407"> </a>GO CHECK IT OUT!!!</li></ul><h3 id="usage-tips-:-tgu2mejds">Usage Tips :</h3><ul><li>The LoRA weight I have been using is 0.7-0.85, but feel free to try and experiment!</li><li>I have used ComfyUI when generated pictures and the settings im using are: Sampler: dpmpp_3m_sdecfg: 3.6 Scheduler: Karras Steps: 40</li><li>For A1111 enjoyers, I would use: Sampler: DPM++ 3M Karras cfg: 3.6 Steps: 40</li><li>Example prompts used:Prompt:"a cinematic shot of a female (ghost spirit:1.3), crying, see through, ectoplasma, sitting on a stone bench in an old Church, crying, hkhorror, masterpiece, best quality, highly detailed, sharp focus, dynamic lighting" Negative prompt:"text, watermark, low quality, worst quality, bad compostion, poor, low effort, child, childish, young, nude, nudity, human"</li><li>First image is with my LoRA and second is without my LoRA, same prompt and model. <img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2ecad85a-7b8c-422b-b78e-3762489b3721/width=525/2ecad85a-7b8c-422b-b78e-3762489b3721.jpeg" /></li><li>Every picture is generated in 1024*1024 with no upscale</li><li>Recomended vae that I use is the official sdxl_vae</li></ul>

Files with SHA256: 8c2b878136f557564344cfc2e4cb6deb77ce190a04d9972e8796e2b75cf88d24