Thread Reader
LudovicCreator

LudovicCreator
@LudovicCreator

May 24, 2024
3 tweets
Tweet

🚨🚨 LEONARDO AI BREAKING 🚨 @Leonardo.Ai double-drop: Image Gen V2 goes live today, and Character Reference is now LIVE for all users To celebrate the below news FREE users get full access to most premium features in the V2 interface until May 30th Image Gen V2 This is a new, simpler interface with a cleaner UI. When you next login to app.leonardo.ai you should see an intro modal welcoming you to the new experience. What you will see (Check the video on main post): Presets: These combine models, styles, elements, and other inputs to achieve specific aesthetics effortlessly. (Coming soon - create and save your own Presets) Simplified settings: Choose your aspect ratio, and we automatically optimize the resolution, based on the model. Easy image guidance options: Style Reference, Content Reference, Character reference easily selectable from the prompt bar. Fast vs quality modes: Toggle between Fast and Quality (formerly Alchemy) Note: If you’re not ready to change, don’t worry, you can still roll with the old interface by using the Legacy Mode toggle in the top right of the interface. Additional note: Increased generation quality for all XL and Lightning models on the platform. It is recommended trying Fast mode on V2 with high resolution ON - check the More section of the image dimensions. Last Image guidance can be found within the little pictuer+ icon just to the left of where you input your prompt. A couple of features are not yet live in V2 but will be there soon, and much more is coming to improve the UI from here. FIRST COMMENT CHARACTER REFERENCE

CHARACTER REFERENCE : With this feature, you can now use a single character image to guide your generations. Character reference works best with human face photos as an input, and may also work with non-human characters. Some example use cases : Film and storytelling: Produce consistent characters for film, TV, graphic novels, and books Fashion design: Apply different outfits and styling to the same model Marketing: Create localised advertising with the same model Gaming assets: Generate consistent characters across different environments, and gear It can be combined with other guidance options like style reference, content reference, and pose to image for additional control It has 3 strength settings: Low strength - greatest flexibility, at the cost of resemblance. Medium strength - reasonable flexibility and more resemblance. High strength - may be less flexible but have the best resemblance.
Mixing Character, Style and Content reference :
LudovicCreator

LudovicCreator

@LudovicCreator
Midjourney Leonardo AI LCP 🎨Prompt Creator🎨 collaborator of @getremixai Chemical engineer sharing prompts with the community @Eternitai_ AI Creator
Follow on 𝕏
Missing some tweets in this thread? Or failed to load images or videos? You can try to .