Thanks for taking a look at the Ray- series of models. Back in November-December 2023, I trained the first model in the series, RAYMNANTS, on a series of thousand-ish personal assets (street photo RAW or edited, digital paintings, etc.) with the goal of creating a model that would make people that look like people, and with the visual style I like best (slightly grainy DSLR photo). From there, I trained two additional family of models with different styles that I link below as well. I kept the models under wraps as my employer was interested in it but since it ended up in a dead-end and with the recent Flux models explosion, I figured I might as well give my work back to the community before SDXL falls into complete obsolescence. Thank you all for the inspiration, knowledge and keeping the dream alive.I hope you folks will have a good time with my models~R. +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Check out my realistic model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/854267">RAYMNANTS</a> Check out my painterly model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/854298">RAYCTIFIER</a> +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ <h1 id="introducing-rayburn1.0-bj5wl7h6c">Introducing RAYBURN1.0</h1>My overly opinionated stylized model, geared comic-style illustrations, it will take everything you thrown at it, and give it back, with a twist.I’ve finetuned a branch of my painterly model, RAYCTIFIER, on a mix of about 800 hand-drawn illustrations and synthetic data.Careful cross-breeding to cultivate some style traits resulted in a very bold style you can’t really prompt out of but that has a lot of personality. It’s seriously overfit, but that’s what makes it the way it is. The model will perform well at taking its own stylistic stance to any kind of prompt you’ll thrown at it. It’s great fun, and also a bit wild! <h2 id="how-to-prompt-v8i41i1h8">HOW TO PROMPT</h2><ul><li>Keep it simple: Just start with subject, then add more details. The model will do its thing. It likes being told colors and will usually try to use them all.</li><li>Use negative only if you need it, the more generic image correction negative you will use the more the output will start looking generic. Furthering this, RAYBURN functions best without generic embeddings.</li></ul>Quick tip: Using higher CFG will result in more “burned” images that would look bad with more realistic content, but work really well with RAYBURN’s illustrative style. Higher CFG gives you flatter, more contrasted colors with bolder shapes, lower CFG gives you more subtle shading and color gradients. You can try to compensate for super high CFG with low steps and vice-versa.<h2 id="examples-sln23zdks">EXAMPLES</h2><ol><li>snow white, saturated, vibrant, colorful</li><li>glitch art,a couch in front of a french doors window, vibrant, aberration, colorful, vaporwave</li><li>Drunk Swedish man, fur hat, wearing a red ski jacket, standing under a snowstorm, pencil strokes, dynamic ink</li></ol><h2 id="settings-kou5usvem">SETTINGS</h2><ul><li>Stay within SDXL picture dimensions for the initial gen if you can. (as a rule of thumb adding height and width together should more or less fall within the 2000 range, i.e. 1024+1024=2048 or 832+1216=2048. 1900 to 2100 is okay, 2150 upwards with probably start repeating latent blocks. Multiples of 64 work better too!)</li><li>The model performs well with a variety of sampler, but I usually just use a classic DPM++ 2M Karras, 20 steps, CFG 2-4. I’ve gotten good results with Heun/AYS too with faster gens at 12-14 steps CFG 4-7.</li><li>I usually do a second pass instead of using an upscaler for the less contrasty stuff, with DPM++ 2M/Karras with anywhere between 4 and 20 steps. For high contrast things, a good old 4xUltrasharp or 4x_NMKDSuperscale_Artisoft_120000_G are good examples. I try to get upscalers that will get me a sharp, crisp line art.</li></ul> <h2 id="known-issuesquirks-6qnne42ob">KNOWN ISSUES/QUIRKS</h2><ul><li>RAYBURN doesn’t listen well to style guidance prompts, but hey, that’s what we’re here for, good luck trying to make it to character photos.</li><li>When generating characters, it has a bit of a tendency to put some strokes or marks on cheeks/forehead. You can prompt out of it, but that can also change your result in general. It’s not necessarily ugly though! Just annoying when you don’t want it.</li><li>RAYBURN loves color, and it tends to add a warmer tone to everything by default (i.e. a prompt for “white snow”  will usually result in a light beige snow.</li></ul>

Files with SHA256: 6ddae056e8df902d676cedba652c063cee063994923fa6bd6225cf85b1abc989