From a spark of passion to a beacon of creativity, Civitai has had quite the year. Step 3. Browse buildings Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsTry adjusting your search or filters to find what you're looking for. it's free. Click on the show extra networks button under the Generate button (purple icon) Go to the Lora tab and refresh if needed. Head to Civitai and filter the models page to “ Motion ” – or download from the direct links in the table above. Bug fix: Now properly works on SD. AI image generator Stable Diffusion already has a lot of fans, and now those experimenting with the new AI technology to develop their own models have a place to share their work with other enthusiasts. Im currently preparing and collecting dataset for SDXL, Its gonna be huge and a monumental task. 1 million round in June led by Andreessen Horowitz (a16z). ai just put through a change to hide large amounts of NSFW content. The model types provide an easy way to set up the configuration for people who’re new to training, but experienced LoRA trainers can tweak. Civitai is the ultimate hub for. . What are depth maps? This guide will walk you through a number of simple techniques designed to bring your AI generated images to life! A depth map is a single channel image that represents the distance of pixels in a scene from the viewer. The GhostMix-V2. AI made the impossible possible :D. Version. Browse pokemon Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsThis resource is intended to reproduce the likeness of a real person. Highest Rated. Our platform allows users to collaborate, share, and access a wide range of AI models specifically designed for art generation. All resources require an associated image post, with at least one image. 5+ Use DPM++ 2M Karras or DPM++ SDE Karras Strength: 0. experiment with the output speed, I mostly use "x3 Speed" with "x2 Slowmo" for smooth 30fps GIFs. In my tests at 512,768 resolution, the good image rate of the Prompts I used before was above 50%. Civitai serves as the exclusive platform for sharing AI-generated art models within the community. AI-generated fake images created in Civitai. Research Model - How to Build Protogen ProtoGen_X3. Terms of ServiceSupport Us ️. Don't intentionally be an asshole, adhere to reddits content. AIモデルのsafetensors形式とは何者か？Haibara Ai LORA Set weight 0. base model anything-v4. The API will return the response status code, response headers, and a response body. It provides free access to over 1,700 models and emphasizes collaboration to create new innovative AI art. Each prompt includes setting, background, pose and (mostly) gender-agnostic character details. Share you models, textual inversions, hypernetworks, aesthetic gradients, and any other crazy stuff people do to customize their AI generations. Browse lora Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs Cost. Browse photorealistic Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs. This command does the following:Browse character Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsFantasy. Check the image to see what prompt i use. Inspiration 7300. V6 Changelog 2023/06/03: Considering this was my first and most popular LoRA, I fig. 4. 0 is suitable for creating icons in. mp4 -i originalVideo. 5D like image generations. Browse civitai Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsNot only is it “under review” on civitai, people have been reporting my huggingface account, I am worried that the samdoesart drama will get cancel my hugging and maybe even my parteon, right now Ai art is almost a part-time job(i’ve been working with a few companies on prompt related stuff) something that hoping to turn fulltime,. You should use this between 0. It allows users to browse, share, and review custom AI art models, providing a space for creators to showcase their work and for users to find inspiration. AI made the impossible possible :D. Updated:. jpeg. Now, we finally have a Civitai SD webui extension!! Update: v1. A big problem with DreamShaper is that it wasn't much compatibles with the LoRA's I was making. Support Us ️. The Unstable Barbie Difference: Stability in Chaos. Support Us ️. Got questions about Stable Diffusion, fine-tuning, or prompting? Dive into our community forum and ask generative AI experts for guidance. Civitai brings a new dimension to AI art, providing a platform for artists to share their models and inspire creativity in others. Terms of ServiceSupport Civitai and get exclusive perks. We're eager to see what the community can produce using pure SDXL! All images submitted must be created after the start of the contest on August 8th. I tried to refine the understanding of the Prompts, Hands and of course the Realism. NOTE: Xpuct's latest "Deliberate v2. AutoV2. I made a doujin style LORA for Haibara Ai before,Now I trained one with the current anime style. A search won't find it, it doesn't appear in the lists of new files or even under NSFW or the Porn categories. The model is the result of various iterations of merge pack combined with. I wanna thank everyone for supporting me so far, and for those that support the creation of SDXL BRA model. 5, as it is more popular, but 2. r/StableDiffusion. posts. Hires steps:17, denoising:0. . Introducing Civitai, the ultimate one-stop model-sharing hub for the AI art generation community! With its unique combination of being free to use, open source, and continuously improving, Civitai makes it easier than ever before to. In these cases, you can always link a model to civitai by filling its URL in this extension. No results found. Update June 28th, added pruned version to V2 and V2 inpainting with VAE. AnimateDiff allows us to inject motion into our txt2img (or. Hash. All Time. fix is needed for prompts where the character is far away in order to make decent images, it drastically improve the quality of face and eyes! Sampler: DPM++ SDE Karras: 20 to 30 steps. Plusle and Minun Pack (Inconsistent) Impidimp Attack. I'm happy to take pull requests. V2. This checkpoint recommends a VAE, download and place it in the VAE folder. Original Hugging Face Repository Simply uploaded by me, all credit goes to . Put the Lora of the first epoch in your prompt (like "<lora:projectname-01:0. 2Dcharacter concept art volumen 1work with 0. Please consider joining my. vae-ft-mse-840000-ema-pruned or kl f8 amime2. So, the file on civitai is actually not the one on your manchine. Log in to view. Version 2, for 2. Next again. Am i Real - Photo Realistic Mix Thank you for all Reviews, Great Trained Model/Great Merge Model/Lora Creator, and Prompt Crafter!!!This resource is intended to reproduce the likeness of a real person. You can go lower than 0. All Time. MeinaMix and the other of Meinas will ALWAYS be FREE. In this article we demonstrated a pipeline for using Civitai models with the diffusers package, from downloading to converting the model, to implementing CLIP skip and prompt embeddings. 5 and 1 weight, depending on your preference. Civitai is a free and open-source platform that aims to foster a community of AI art enthusiasts and creators. Us ️. Browse tifa Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsCivitai, an online marketplace for AI models, has recently introduced a controversial feature allowing users to post "bounties" for creating deepfake images of real people, including normal people without significant public presence. Use aihaibara to summon her, (you may add child to guide AI to generate her younger, but the model already generated her very young) Use short hair, brown hair, bangs, hair between eyes, blue eyes to form her head, you can get creative on her outfit tag. Open detection tab. New to AI image generation in the last 24 hours--installed Automatic1111/Stable Diffusion yesterday and don't even know if I'm saying that right. Out of respect for this individual and in accordance with our Content Rules, only work-safe images and non-commercial use is permitted. Begining A List of every pokemon creature model in the site, let me know if there is something missing from the list. bounties. SDXL Generations require a minimum of 3 Buzz , for a single 30 step image. Counterfeit-V3 (which has 2. The change in quality is less than 1 percent, and we went from 7 GB to 2 GB. CivitAi’s UI is far better for that average person to start engaging with AI. The pursuit of perfect balance between realism and anime, a semi-realistic model aimed to ach. waeys. Bolstered by a $5. Civitai stands as the singular model-sharing hub within the AI art generation community. Civitai is the ultimate hub for. Andreessen Horowitz, also known as a16z, the influential Silicon Valley venture capital firm that was an early investor in Facebook, Lyft, and other tech giants,. but the CivitAI team consists of 4 people, and apparently it's costing them around $4,000 USD to run and operate (server costs and. Natural Sin Final and last of epiCRealism. . Browse from thousands of free Stable Diffusion models, spanning unique anime art styles, immersive 3D renders, stunning photorealism, and more. 9 I just use few pictures to train this lora, but the results seem pretty good. 2023/3/3 update Update dataset and model. Eveevolutions + Shaymin + Absol Pack (YiffyMix) Vaporunny (Vaporeon x Lopunny Fusion) Muscular Pikachu. All Time. Huge spam of calls to Instagram API & CDN. x after v1. 5. Verson2. 5 and 1 weight, depending on your preference. x. Unique nameplate color. VisionGen v1. ComfyUI is a super powerful node-based, modular, interface for Stable Diffusion. Browse comic book Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs. base model :. Charturner keeps the outfit consistent, controlNet openPose keeps the turns under control. After weeks in the making, I have a much improved model. fix: R-ESRGAN 4x+ | Steps: 10 | Denoising: 0. Sinkin. Browse photorealistic Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs. Browse text Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsSensitive Content. Hires. 5 You guys can support me on Patreon. This is still in active development and will be updated once more endpoints are made available for the public. Click "show more" to view a full list of features and install guide below! ↓ ↓ ↓ ↓ ↓ Features 🚀 Browse all models from CivitAI 🧩. . Pruned SafeTensor. This is a general purpose model able to do pretty much anything decently well from realistic to anime to backgrounds All the images are raw outputs. 404Media points to another more serious problem Civitai has — it allowed users to post non-consensual porn images, and created a bounties feature for making AI images of certain targets, styles. Out of respect for this individual and in accordance with our Content Rules, only work-safe images and non-commercial use is permitted. How to Get Started with Civitai! This guide focuses on the Basics - Creating & Configuring an account, navigating the Home Page, & Discovering content. The "MagicAI" is not meant to be a trigger word. This step is crucial for accurate training and better results. It works better with my character Lora's than even the base SDXL model (usually moving away from the base model lessens likeness). 1. The model types provide an easy way to set up the configuration for people who’re new to training, but experienced LoRA trainers can tweak. Whether you’re an experienced artist looking to explore new avenues or a newcomer eager to dive into this exciting. This model is an extremely robust general purpose anime model,. Feature Request. There is a wide variety of choices on the Civitai site, and manga style has a good place. Enhance character specific eyes (Hu Tao, AI Hoshino, etc. From here, the Training Wizard begins, starting with the initial page – the Model Type Selection; Choosing a model type pre-sets some of the advanced training settings. Browse semi-realistic Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic. - GitHub - BlafKing/sd-civitai-browser-plus: Extension to access CivitAI via WebUI: download, delete, scan for updates, list installed models, assign tags, and boost downloads with multi-threading. At Civitai AI, we believe in empowering artists by harnessing the power of artificial intelligence. The model's owner changed file on civitai, but does not change version name and description. ️. Sampler: DPM++ SDE Karras, 20 to 30 steps. Browse concept Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsExplore the experience of creating Stable Diffusion AI Art with Civitai. Others. Submit your Part 2 Fusion images here, for a chance to win $5,000 in prizes!Checkpoints and Loras used in this mix: AbyssHellMapple Toonify v1 adjusting your search or filters to find what you're looking for. Explaining the Civitai. An AI can easily be trained to generate a wide range of styles, from photorealistic images to more abstract designs. 5 and "Juggernaut Aftermath"? I actually announced that I would not release another version for SD 1. In all the images I posted I was using Dynamic Thresholding at 3. Civitai is a website where you can browse and download lots of Stable Diffusion models and embeddings. 0. Updated daily. What makes Civitai truly stand out from the crowd, though, is its emphasis on community. This is a Wildcard collection, it requires an additional extension in Automatic 1111 to work. The model's owner changed file on civitai, but does not change version name and description. Sizes: 512x512, 512x768, 512x904 (Can be used both vertically and horizontally). Pruned SafeTensor. With its intuitive interface, advanced search features, and huge range of available models, Civitai makes it simple to find, modify, and share the best AI models for your needs. Browse pixel art Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsstyle cyber punk film photorealistic sexy female. Version 2, for 2. Civitai is a free & open source AI model-sharing hub for the AI art generation community where users can browse and download lots of Stable Diffusion models and. Civitai Guide to Buzz – the Civitai CurrencyTry adjusting your search or filters to find what you're looking for. . style cyber punk film photorealistic sexy female. mayonnaise(?) Perpetual coffee request: I signed up for Buy-Me-a-Coffee and Ko-fi, mainly because I thought it would be hilarious if someone as artistically untalented as myself received a coin for something creative. 5. FFUSION AI is a state-of-the-art image generation and transformation tool, developed around the leading Latent Diffusion Model. Most of the sample images follow this format. Explore thousands of high-quality Stable Diffusion models, share your AI-generated art, and engage with a vibrant community of creatorsBrowse 30 Civitai AI tools. These generated images are for personal usage only. Civitai AI: A platform for sharing AI-generated art Welcome to Civitai AI, the premier destination for artists and enthusiasts in the AI art generation community. each month. This is by far the best SDXL checkpoint I have used. 3 After selecting SD Upscale at the bottom, tile overlap 64, scale factor2. its free. Civitai上では沢山の人がモデルやLoRAを公開しており、それらのデータを無料でダウンロードできる神サイトです。 本記事では、超有益サイトであるCivitaiの使い方をAI画像生成初心者でもわかるように徹底解説します。 Browse japanese Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs Super Easy AI Installer Tool. AI Hands Prompts to start with : --Name-- showing hands Trained using SDXL trainer. Hoshino Ai, long hair, purple hair, streaked hair ,purple eyes, star-shaped pupils, hair ornament, model: most of anime-like models. Comprehensive database of AIs available for any use case. Discord. Try adjusting your search or filters to find what you're looking for. FFUSION AI is a state-of-the-art image generation and transformation tool, developed around the leading Latent Diffusion Model. Subscribe. ChikMix_v3 now release! Blend with braBeautifulRealistic_v40 Feature: 2. Try adjusting your search or filters to find what you're looking for. VAE recommended: sd-vae-ft-mse-original. Realistic. Make awesome images! Note: You can use a LoRA with. Join Us 💼. Filters. Feature Request. JAV Ai_佐山 557. + 6. Share you models, textual inversions, hypernetworks, aesthetic gradients, and any other crazy stuff people do to customize their AI generations. Aboslutley beautiful checkpoint and i'm having a lot of fun using it. [Step 3: Tagging Images] Once you have your images, use a tagger script to tag them at 70% certainty, appending the new tags to the existing ones. Unique Supporter Tier badge. Created as described in the tutorial: Perpet. Place the file inside the models/lora folder. But the models are there because their authors want them to be public. ai. 请避免使用例如只有“1girl”此类的提示词，模型提示词准确度极高， 对使用者的prompt书写能力会有所要求. We are going to be describing the HTTP method, path, and parameters for every operation. 0. Kind of generations: Fantasy. (Sorry for the misunderstanding caused by. What is Stable Video Diffusion (SVD)? Stable Video Diffusion (SVD) from Stability AI, is an extremely powerful image-to-video model, which accepts an image. Civitai is a website where you can browse and download lots of Stable Diffusion models and embeddings. Civitai is an online platform for AI-generated art models using Stable Diffusion and a collaborative artist community. CFG Scale: 7. m. Browse pixel art Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs More attention on shades and backgrounds compared with former models ( Andromeda-Mix | Stable Diffusion Checkpoint | Civitai) Hands-fix is still waiting to be improved. To reproduce my results you MIGHT have to change these settings: Set "Do not make DPM++ SDE deterministic across different batch sizes. It makes objects made of ivory and gold. It makes objects made of ivory and gold. + 6. Limited time supporter option. 3 includes improvements from Version 5, offering slightly enhanced brightness; however, its performance remains subpar when it comes to hair ornament details. 0. See the latest art created by the generative AI art community and delve into the inspirations and prompts behind their work. From here, the Training Wizard begins, starting with the initial page – the Model Type Selection; Choosing a model type pre-sets some of the advanced training settings. This compares to $20 million for companies that have fine-tuned an open-source model, and $9 million for “wrapper” companies. 0 significantly improves the realism of faces and also greatly increases the good image rate. Thank you thank you thank you. Civitai hosts thousands of models from a growing number of creators, making it a hub for AI art enthusiasts. py. 2. (如果你第一次接触AI绘画，我建议你使用v16) （喜欢哪个模型用哪个）-----AND Please add AI tag if you post images on media! (The images generated by AI are currently in a gray area. Big update of DucHaitenAIart, v4 is able to receive more diverse, more detailed prompts with gorgeous colors and more realistic shadows. All Time. Anything-Ink. . Not like civitai owns anything at all. Gives dragon scales and spikes on things. 266,651 searches today / The #1 AI aggregator. Suggested by memesis Created as described here: . Try adjusting your search or filters to find what you're looking for. there I do fan request model training. Browse anime character Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAs. Multipurpose, all-in-one AI app to generate text, image, code, story, poem, and to analyze image and text, and much more. Join us as we redefine the future of artistic expression!right now they are pretty much the best resource for anything A. . ChatGPT has a massive lead, for now…. 8346 models. Browse poses Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsTry adjusting your search or filters to find what you're looking for. 8 - Rebuilt - How to Build Protogen By Downloading you agree to the following licenses CreativeML Open RAIL-M Dreamlike License. Civitai said that it raised $5. Tons of models there are on other places like hugginface too. Join Us 💼. 5. So I wanted to create a model that was "artistic" like DreamShaper but also able to use my. Acceptable file types for static images are . Through this process, I hope not only to gain a deeper. A model for large breasted waifus or semi-realistic characters. Plan and track work Discussions. Creators earn money for completing the bounties, which may be used for deepfake porn or other nefarious purposes. Highest Rated. I have a brief overview of what it is and does here. mp4 -map 0:v -map 1:a -c:v copy -c:a aac output. 5. caption. This extension provides assistance in installing and managing custom nodes for ComfyUI. It's still there, you just can't find it unless you have the link. Try adjusting your search or filters to find what you're looking for. Terms of ServiceTry adjusting your search or filters to find what you're looking for. Bad Dream + Unrealistic Dream (Negative Embeddings, make sure to grab BOTH) Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee ☕. This technology is being used in a variety of applications, including chatbots, content creation, and even computer programming. You can now run this model on RandomSeed and SinkIn . This commitment ensures that Civitai remains at the cutting edge of AI art generation, providing a hub that meets the changing needs of the community it serves. articles. Nov 11, 2023. Hires upscale: The only limit is your GPU (I upscale 2,5 times the base image, 576x1024) VAE:. Step 2. Hope you like this lora, and anime too. Model type: Diffusion-based text-to-image generative model. x after v1. 111 upvotes · 20 comments. Remember to use the pruned version when training (less vram needed). Announcement: Versions prior to V0. Since SDXL is right around the corner, let's say it is the final version for now since I put a lot effort into it and probably cannot do much more. #community #database #download #dreambooth. Use the. But the models are there because their authors want them to be public. webm. Try adjusting your search or filters to find what you're looking for. Link to the Civitai animatediff guide in bio. CivitAi’s UI is far better for that average person to start engaging with AI. This step is crucial for accurate training and better results. If you're unsure of what version you are running, it's probably 1. 2. . If you are the person or a legal representative of the person depicted, and would like to request the removal of this resource, you can do so here. 0-locon | Stable Diffusion LyCORIS | Civitai. It works well on dreamshaper_4BakedVae. Not like civitai owns anything at all. They are hosting loads of enormous models and images for free, on a very spiffy, heavy website. These files are Custom Workflows for ComfyUI. It needs to be in this directory tree because it uses relative paths to copy things around. AnimateDiff in ComfyUI is an amazing way to generate AI Videos. Character commission is open on Patreon Join my New Discord Server Use the preview. There are recurring quality prompts. These LORAs were mixed within the model using different weights. Step 4 – Create a post. 5. img2img SD upscale method: scale 20-25, denoising 0. ; GPT-3/4 (Chat GPT) is another example of generative AI – a language model that can generate human-like text. It’s GitHub for AI. No results found. 40. 7 weight enjoy it one of my first loras and the ones I liked the most I will divide it into 3 volumes for greater use, recomend use hirex fix and impating for details,flexible for lora character, resolution 768x768, 512x768. 5, as it is more popular, but 2. 5D like image generations. civitai is a platform for stable diffusion ai art models. It. 5~0. 85 a good starting value and adjust as needed. Set Model 1 to the LoRA model you downloaded. Our. Explore thousands of high-quality Stable Diffusion models, share your AI-generated art, and engage with a vibrant community of creators. Civitai is an AI image generator using Stable Diffusion and a community of artists through an online platform. Browse anime Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsIt’s GitHub for AI. x. . ChatGPT optimizes. Learn how to navigate the world of generative AI with Civitai, the largest platform for discovering, creating, and sharing generative art. Sampler: DPM++ 2M Karras, 20 to 40 steps. All my new SDXL Loras are available for free to use as early access in Tensor. If civitai makes too much of a barrier to entry, the site will die, but its content will still be available. Sampler: DPM++ SDE Karras, 20 to 30 steps. ai just put through a change to hide large amounts of NSFW content. What kind of dumb comments are these lol. Place the file inside the models/lora folder. They have asked that all i. Category : Art. Swole Torchic. Size: 512x768 or 768x512. Opening Civitai and idling at the homepage generate hundreds of calls to the Instagram API & CDN in a short span of time. It’s often used to create 3D images or models from 2D images, and provide information about. A111 -> extensions -> sd-civitai-browser -> scripts -> civitai-api. Explore thousands of high-quality Stable Diffusion models, share your AI-generated art, and engage with a vibrant community of creatorsThis is my complete guide how to Generate sprites for 8bit games or GIFs :) Enjoy the video Use it with my toolkit to get similer results to the on. Terms of ServiceSupport Us ️. It allows you to browse CivitAI, Check for updates and speed up downloads. It was trained on v1-5-pruned for 5000 steps. civitai 알아보기. From using generative AI features and collaborating with builders to tipping fellow Civitans and personalizing your Civitai profile, Buzz can now be used in a variety of useful ways on Civitai. Crystalline thingies. The preview pictures were generated with prompt: <lora:ElementalMagicAIv2-000008:0. No results found. This is a builtin feature in webui. Highres-fix (upscaler) is strongly recommended (using the SwinIR_4x,R-ESRGAN 4x+anime6B by myself) in. A search won't find it, it doesn't appear in the lists of new files or even under NSFW or the Porn categories. Discover an imaginative landscape where ideas come to life in vibrant, surreal visuals. The “Style Asian Less” model, which has been downloaded more than 2,000 times on the AI art community site Civitai, can be applied to the open-source AI image generating tool Stable Diffusion. 3 up to 0. A repository of models, textual inversions, and more - Home · civitai/civitai Wiki. Our application empowers you to generate precise text caption files for each image using an intuitive UI and combining your insights with optional AI assistance for enhanced language quality. 2 will no longer dete. Use the preview images as a starting pointBrowse peeing Stable Diffusion models, checkpoints, hypernetworks, textual inversions, embeddings, Aesthetic Gradients, and LORAsYou can run XXMix_9realistic on sinkin. 5D K-doll style focus Super simple prompts. I also found out that this gives some interesting results at negative weight, sometimes. It is capable of completing sentences, paragraphs, and even entire articles, given a short prompt. I also found out that this gives some interesting results at negative weight, sometimes. AI Hands Prompts to start with : --Name-- showing hands Trained using SDXL trainer. I've been making LoRA's non stop for the last month, but I felt like I needed something. All Time. This model would not have come out without XpucT's help, which made Deliberate. Use the negative prompt: "grid" to improve some maps, or use the gridless version. 45 | Upscale x 2.