- Civitai Newsletter
- Posts
- New content discovery options, base models, Artist-in-Residence update, and more!
New content discovery options, base models, Artist-in-Residence update, and more!
Civitai | Discord | Feedback | Bug Report | Merch
Greetings, Civitans! Welcome to the twenty-sixth edition of Civitai Biweekly, your essential guide to all recent developments and updates within Civitai.
This week, we’ll cover:
Model and Image Discovery updates! 🔍
New filtering options! ⚗️
New troubleshooting features! 🔧
Artist-in-Residence Updates! 🎨
…and more!
Civitai Developments
1️⃣ Model & Image Filtering and Discovery Updates
Discovering historic content is now a lot easier! We’ve implemented an Oldest filter, available for both Images and Models - it’s Newest, in reverse!
There’s also a new Made On-Site filter for Models and Images, which, when selected, will narrow down results to display only the wonderful content generated by our On-Site LoRA trainer and Image Generator!
2️⃣ New Base Model Types
We’ve added Base/Foundational model support for four new model types;
Pony Diffusion v6 XL (sufficiently different from SDXL to warrant its own type!)
SDXL Lightning
Stable Cascade
New Base Model Filters
3️⃣ Licensing and Commercial Use Settings Update
The Licensing and Commercial Use settings have received a much-needed overhaul moving away from the “most permissive” slider to a much more granular checkbox style implementation.
We’re currently updating our documentation and will have the full Licensing Guide available on the Education Hub soon!
Licensing and Commercial Use Settings - now with checkboxes!
4️⃣ New Generator Troubleshooting Features
Our Image Generator now boasts an invaluable feature: the ability to effortlessly copy Image Generation Job IDs for the most recent batch of images. When encountering any issues with the Generator, simply click the designated button to copy the Job IDs directly to your clipboard.
By pasting these IDs into any communications you have with our support team, you’ll significantly aid our Developers in diagnosing and troubleshooting Generation issues!
Contests
Our dual Valentine’s day and Chinese New Year contests have ended! We want to thank everyone who participated! We had over 3000 images and 100 resources submitted and there were so many amazing entries it made it truly hard for us to do staff picks!❤️
You can view all the entries in the following collections;
Our winning images across all Image categories;
Valentine reaction winner - FallenIncursio | Valentine collection winner - okamuron | Valentine Civitai Staff Pick winner - MGod |
Lunar reaction winner - smth_uncanny | Lunar collection winner - nemosenchou798 | Lunar Civitai Staff Pick winner - OneViolentGentleman |
Everyone who submitted an entry to at least one of the Contest Collections will now have received the beautiful Lunar Love Contest Badge!
Generative AI News
1️⃣ Stable Diffusion 3
Stability AI have announced their next evolution of Generative AI txt-2-img models; Stable Diffusion 3.
Prompt: Epic anime artwork of a wizard atop a mountain at night casting a cosmic spell into the dark sky that says "Stable Diffusion 3" made out of colorful energy
Claimed to be Stability’s most capable txt2img model, it promises; “greatly improved performance in multi-subject prompts, image quality, and spelling abilities.”
It’s only available via early-access preview right now, and you can sign up for the waitlist here.
2️⃣ New LoRA Technology - DoRA!
Last week, researchers proposed exciting improvements to the LoRA training process in the paper; Weight-Decomposed Low-Rank Adaptation (DoRA). DoRA trained models;
“…apply the decomposition into magnitude and directional components to a whole pretrained weight matrix W instead of a vector, where each column (vector) of the weight matrix corresponds to the weights connecting all inputs to a particular output neuron.” - Sebastian Raschka, PhD
Simple!
Guides & Tutorials
1️⃣ YouTube Tutorial Series!
We’re continuing to expand our YouTube Beginner’s Guide to Generative AI Art tutorial series! If you’re a newcomer to the AI art scene and want to absorb all the essential info in the shortest time possible, our concise video guides are the the resource you need!
Artist in Residence
We are thrilled to announce the inaugural class of artists selected for Civitai’s new Artist in Residence Program! This diverse group brings a rich blend of creativity, technological innovation, and conceptual inspiration to our community.
We look forward to sharing their progress and insights throughout their residency in the STUDIO section of our site, where each artist will share their individual creative process. Join us in welcoming Noah, Domenec, Arvid, and Shinji to the Civitai community.
Resource Showcase
Evil Animals LoRA for SDXL
Unleash the enigmatic and eerie. Inspired by neo-traditional tattoo art, the Evil Animals LoRA from Creator Lara_De_Martin conjures creatures that blur the line between fascination and fear, producing images with a darkly distinctive style.
What’s on Your Mind LoRA for SDXL
An experiment in unfiltered cognition from Creator Dever! Pixels meet profundity with the beautifully crafted What’s on Your Mind LoRA.
Trending Images
Thanks for sticking with us all the way to the end! Here’s your well-earned Newsletter Buzz ⚡ - consider it a ‘thank you’ for being a valued, informed, member of the Civitai community!
Office Hours
Want to talk directly with the team behind Civitai and learn more about the website and generative AI?
Join our Office Hours on Discord, and Twitch where community members of all skill levels come together to ask questions, delve into insightful discussions, and share best practices for generative AI! We look forward to seeing you there 👋