<h1 id="ebe-imagine-diffusion-mg9ru2etd"><a rel="ugc" href="https://huggingface.co/enhanced-by-ehristoforu">EBE</a> - imagine diffusion</h1>EBE (Enhanced By Ehristoforu) is a modern project whose task is to create the best universal and balanced diffusion models based on SD1.5, SD2.1, SDXL, PonyV6, SD3 and others.<h2 id="about-this-project-9ll4goesr">About this project</h2><h3 id="who-is-working-on-the-project-huay7bhld">Who is working on the project</h3>The Project Fluently and DreamDrop team led by ehristoforu are working on this project.<h3 id="what-are-the-goals-of-the-project-hvwenthl3">What are the goals of the project</h3>Our project has several key goals:<ul><li>Creating the most versatile and balanced models</li><li>Creating a friendly community in Discord/HuggingFace, where everyone will participate in improving the models</li><li>Creation of LoRAs for different tasks</li><li>Creating styles for our models</li></ul><h3 id="current-roadmap-izq2q34mq">Current roadmap</h3>Our roadmap looks like this:<ol><li>Creation of EBE based on SD1.5</li><li>Creation of EBE based on SD2.1</li><li>Creation of EBE based on SDXL</li><li>Creation of EBE based on PonyV6</li><li>Creation of EBE based on SD3 Medium</li><li>Creation of LoRAS for EBE based on SD1.5</li><li>Creation of LoRAS for EBE based on SDXL</li><li>Creation of LoRAS for EBE based on PonyV6</li><li>Creation of LoRAS for EBE based on SD3 Medium</li><li>Create a large number of quality styles for EBE models</li></ol><h2 id="how-our-models-are-created-x29i9a2d6">How our models are created</h2>We use several techniques at once:<ul><li>Training base models on large datasets</li><li>Megring well-selected models</li><li>We train LoRAs and integrate them into based models</li></ul><h3 id="is-merging-harmful-to-the-model-ol249n2fc">Is merging harmful to the model?</h3>In our case no, we carefully select and test models for merge, run tests using X/Y/Z plot and much more, so don't worry.<h2 id="links-jqq18pe2w">Links</h2>Links to our resources:<ul><li>Our LoRAs: coming soon</li><li>Our styles: coming soon</li><li>HuggingFace: <a target="_blank" rel="ugc" href="https://huggingface.co/enhanced-by-ehristoforu">here</a></li></ul><ul><li>Our site: coming soon</li><li>Our Discord community: coming soon</li></ul><h3 id="thanks-ds2iklc2s">Thanks 💖</h3><ul><li><a rel="ugc" href="https://civarchive.com/user/KandooAI">KandooAI</a></li><li><a rel="ugc" href="https://huggingface.co/XpucT">XpucT</a></li><li><a rel="ugc" href="https://civarchive.com/user/Zovya">Zovya</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/SG_161222">SG_161222</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/PurpleSmartAI">PurpleSmartAI</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/epinikion">epinikion</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/fp16_guy">fp16_guy</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/Yamer">Yamer</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/Samael1976">Samael1976</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/socalguitarist">socalguitarist</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/PromptSharingSamaritan">PromptSharingSamaritan</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/AreThoseLevel4Plates">AreThoseLevel4Plates</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/olivetty">olivetty</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/Afroman4peace">Afroman4peace</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/RunDiffusion">RunDiffusion</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/buffert">buffert</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/freek22">freek22</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/sinatra">sinatra</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/Corcelio">Corcelio</a></li><li><a target="_blank" rel="ugc" href="https://civarchive.com/user/DataVoid">DataVoid</a></li></ul>

EBE - imagine diffusion

Work done:<ul><li>The model was trained on a large dataset.</li><li>More than 20 best models took part in the merger</li></ul>Results:<ul><li>Improved overall aesthetics and quality</li><li>Improved understanding and adherence to prompts</li><li>Improved tokens: cinematic, depth of field, semi-realistic</li></ul>

Files with SHA256: 6f278d7bf55fa0511a34adb3a28a7c10265125b70c1a21cab8e6e09c6140d869