Got the perfect background but missing your character? Magic Fill makes adding your character seamless. Upload your image, pick Magic Fill, mark where your character should go, add the reference image, and generate — your character appears perfectly integrated.
CREATIVE AI
-
Ideogram Character: Single Reference Image Consistency Model
By
–
Meet Ideogram Character — the very first character consistency model that needs only a single reference image. Now free for all users!
— Futurepedia – Learn to Leverage AI (@futurepedia_io) 30 juillet 2025
Generate consistent character images in any style, pose, expression, scene, or lighting you want. pic.twitter.com/sZ7GGZl6dsMeet Ideogram Character — the very first character consistency model that needs only a single reference image. Now free for all users! Generate consistent character images in any style, pose, expression, scene, or lighting you want.
-
AI Character Generator with Magic Fill and Template Variations
By
–
Just upload a photo or snap a selfie, type your prompt, and hit generate. Or use our ready-made templates to craft stunning variations of your character effortlessly.
— Futurepedia – Learn to Leverage AI (@futurepedia_io) 30 juillet 2025
Plus, place your character into any scene or meme instantly with Magic Fill. pic.twitter.com/nd8XpfrYeVJust upload a photo or snap a selfie, type your prompt, and hit generate. Or use our ready-made templates to craft stunning variations of your character effortlessly. Plus, place your character into any scene or meme instantly with Magic Fill.
-
Generative AI as a complete video production studio
By
–
Attends… Tu donnes juste une idée, et l’IA te crache une vidéo complète avec voix, montage, et tout le reste ? C’est plus qu’un co-pilote, c’est carrément un studio dans ta poche.
-
Reinforcement Learning Enhances Discrete Autoregressive Image Generation Models
By
–
X-Omni Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again
-
HunyuanWorld 1.0: Generate Interactive 3D Worlds from Text or Images
By
–
HunyuanWorld 1.0
— AK (@_akhaliq) 30 juillet 2025
Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels pic.twitter.com/JS7TM2cOgoHunyuanWorld 1.0 Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels
-
Wan 2.2 Video Model Now Faster and Cheaper with Pruna AI
By
–
We've worked with Pruna AI again to make Wan 2.2 faster, and cheaper.
— Replicate (@replicate) 30 juillet 2025
33 seconds, $0.06 per video, this is the large 14b model at 480p.https://t.co/eE3wWPMwLF 💨
Wan more? pic.twitter.com/SXtCYn3uiEWe've worked with Pruna AI again to make Wan 2.2 faster, and cheaper. 33 seconds, $0.06 per video, this is the large 14b model at 480p. https://
replicate.com/wan-video/wan-
2.2-t2v-480p-fast
… Wan more? -
Moonvalley Launches Sketch-to-Video in Marey Model
By
–
Moonvalley announced sketch-to-video, an emergent capability in its Marey video model
— The Rundown AI (@TheRundownAI) 30 juillet 2025
Using only text and a rough hand sketch as input, it creates a ready-to-use, realistic clippic.twitter.com/8vvfoiA8EWMoonvalley announced sketch-to-video, an emergent capability in its Marey video model Using only text and a rough hand sketch as input, it creates a ready-to-use, realistic clip
-
Ideogram Releases Character Model for Consistent AI Image Generation
By
–
Ideogram just released Character, a new AI model for character consistency
— The Rundown AI (@TheRundownAI) 30 juillet 2025
Using just one reference image, it creates consistent character images in any style, expression, scene, and lighting
Available to all Ideogram web and iOS users, for freepic.twitter.com/ZcaopSggovIdeogram just released Character, a new AI model for character consistency Using just one reference image, it creates consistent character images in any style, expression, scene, and lighting Available to all Ideogram web and iOS users, for free
-
Google NotebookLM Adds Video Overviews with AI Narration
By
–
Google's NotebookLM introduced a revamped UI with Video Overviews
— The Rundown AI (@TheRundownAI) 30 juillet 2025
It allows users to convert source materials into video slides, complete with narration.
The update also includes the ability to create multiple outputs from the same sourcespic.twitter.com/NVjOTEWoDEGoogle's NotebookLM introduced a revamped UI with Overviews It allows users to convert source materials into video slides, complete with narration. The update also includes the ability to create multiple outputs from the same sources