<h3>Update 2:</h3>ReV_3 released. Read model version description for more info<h3>Update</h3>ReV_2 released. Read model version description for more info Intro This model is mainly intended for Portraits and Full Body (Depth of Field) like semi-photorealistic pictures. This can do landscape however, not well. I like this model, jack of all trades, master of none. No restrictions on use of this model, just credit to me and other model creators used in this merge.<h3>Workflow</h3>minimal workflow logic based on Protogen's: <a target="_blank" rel="ugc" href="https://www.reddit.com/r/StableDiffusion/comments/1079c0d/protogen_checkpoint_merging_data_reference/">Reddit</a>(Mainly used Difference sum from SD1.5-pruned as tertiary model) <h3> Recommends</h3>Resolution:<ul><li>2:3 ratio (i.e 512x768)</li><li>1:1 ratio (i.e 512x512 or 576x576)</li></ul>Best Samplers:<ul><li>Euler a (Recommended)</li><li>DPM++ 2M Karras</li><li>DPM++ SDE Karras</li><li>PLMS</li></ul>CFG scale:<ul><li>Portraits: 7-9</li><li>Landscape: 9-12</li></ul>Steps:<ul><li>Portraits: 30-40</li><li>Landscape: 40-100</li></ul>Hires.fix: <ul><li>Upscaler: ESRGAN_4x or ESRGAN_4x+</li><li>Upscale by: 1.5-2</li><li>Denoising strength: 0.3-0.5 (0.3 used in samples)</li><li>Clip: 1</li></ul>For best results use the: <a target="_blank" rel="ugc" href="https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main">Standard SD VAE</a><a target="_blank" rel="ugc" href="https://drive.google.com/file/d/1kJsQ4GSoPDfuworuajvTo4xIS-GJUkkX/view">bad-image-prompt-v2</a>SFW: put 'NSFW' and 'uncensored' in negative prompt, SFW in regular promptNSFW: put 'SFW' and 'censored' in negative prompt, NSFW in regular prompt Less negative prompts the better<h3> License</h3><a target="_blank" rel="ugc" href="https://huggingface.co/spaces/CompVis/stable-diffusion-license">creativeml-openrail-m</a>

REV Mix

Major changes: Core model: Deliberate + (f222 - <a target="_blank" rel="ugc" href="https://huggingface.co/runwayml/stable-diffusion-v1-5">v1.5-pruned-emaonly</a>) @ 0.25MBW = SmoothStep*2 derivative Similar Realistic version compared to V2 however, this is more capable of doing more fantasy/sci-fi related prompts. Landscape is better as long you're explicit in your prompts.For Female Portraits try booru tags: ((masterpiece, best quality)) in beginning of prompts Major models used<a target="_blank" rel="ugc" href="https://civitai.com/models/2661/uber-realistic-porn-merge-urpm">URPM v1.2</a><a target="_blank" rel="ugc" href="https://civitai.com/models/5263/rfktrs-darkdream-diffusion">RFKTR's Darkdream Diffusion</a><a target="_blank" rel="ugc" href="https://civitai.com/models/1188/f222">F222</a><a target="_blank" rel="ugc" href="https://civitai.com/models/4823/deliberate">Deliberate</a><a target="_blank" rel="ugc" href="https://civitai.com/models/1116/rpg">RPG v4</a><a target="_blank" rel="ugc" href="https://huggingface.co/cafeai/cafe-instagram-sd-1-5-v6">Cafe Instagram Unofficial Test v2</a>

Files with SHA256: fe39bfd93c368abbb806e9342b89ed58f33440fce9ebc0b950bf02079e3a862a