For whatever its worth, V7 is strictly better at fingers than V6 in all our side-by-side tests and we just released a new update today that improves things even further
MULTIMODAL AI
-
V7 Model Update: Enhanced Image Quality and Creative Aesthetics
By
–
3 quick updates! We've updated the V7 model with improved image quality and coherence. There's a new lightbox editor that's easier to use. Last we've added a new experimental aesthetics parameter –exp (goes 0 to 100, 0 is default) that pumps up details and creativity. Have fun!
-
Character.AI Launches AvatarFX Feature to Premium Users
By
–
AvatarFX has officially rolled out to all c.ai+ users — and the creations you’re sharing? Nothing short of masterpieces. #avatarfx #characterai
— Character.AI (@character_ai) 30 avril 2025
🦞 Mob Boss Lobster
Props to Character.AI user WeeklyBots for this underworld undersea icon.
1/3 pic.twitter.com/pNaM4diQ2wAvatarFX has officially rolled out to all c.ai+ users — and the creations you’re sharing? Nothing short of masterpieces. #avatarfx #characterai Mob Boss Lobster
Props to Character.AI user WeeklyBots for this underworld undersea icon. 1/3 -
Gen-4 References Breakthrough: Enhanced Control and Continuity Features
By
–
Gen-4 References is a major breakthrough for adding control and continuity to your projects. We will continue to add new features and improvements to References, including more robust support for styles, objects and more.
— Runway (@runwayml) 30 avril 2025
(3/4) pic.twitter.com/JijowXTIErGen-4 References is a major breakthrough for adding control and continuity to your projects. We will continue to add new features and improvements to References, including more robust support for styles, objects and more. (3/4)
-
Gen-4 References Rolling Out for Character and Style Applications
By
–
While Gen-4 References currently works best with characters and locations, you can also experiment with different use cases like reference styles. Simply provide an image of your character or location and an image of the style you’d like to apply.
— Runway (@runwayml) 30 avril 2025
References is now rolling out… pic.twitter.com/uSMUGFfz9qWhile Gen-4 References currently works best with characters and locations, you can also experiment with different use cases like reference styles. Simply provide an image of your character or location and an image of the style you’d like to apply. References is now rolling out
-
Gen-4 References Enables Consistent Character and Location Generation
By
–
Today we are releasing Gen-4 References to all paid plans. Now anyone can generate consistent characters, locations and more. With References, you can use photos, generated images, 3D models or selfies to place yourself or others into any scene you can imagine.
— Runway (@runwayml) 30 avril 2025
More examples… pic.twitter.com/Je59RsRnyWToday we are releasing Gen-4 References to all paid plans. Now anyone can generate consistent characters, locations and more. With References, you can use photos, generated images, 3D models or selfies to place yourself or others into any scene you can imagine. More examples
-
AI Image Generation: Place Characters in Custom Scenes
By
–
Combine a single character reference image with any location reference image to place your character in that scene. Prompt for new camera angles, compositions, character styling and more.
— Runway (@runwayml) 30 avril 2025
(2/4) pic.twitter.com/haI0G65e4FCombine a single character reference image with any location reference image to place your character in that scene. Prompt for new camera angles, compositions, character styling and more. (2/4)
-
FAST Framework Boosts Multimodal LLM Accuracy While Reducing Tokens
By
–
Multimodal LLMs often "overthink"—producing long, verbose answers even for simple visual questions FAST (Fast-Slow Thinking for LVLMs) achieves SOTA accuracy while slashing token usage +10% accuracy over baselines Up to 67% fewer tokens used Trending on alphaXiv
-
GPT-4o Converts Whiteboard Sketches Into Professional Visuals
By
–
7. Transformez des croquis sur tableau blanc en visuels professionnels Avec GPT-4o, il est désormais possible de convertir des esquisses rapides faites sur un tableau blanc en graphismes soignés, alignés avec votre identité visuelle — le tout en quelques secondes.
-
Digital Illustration Prompt Using Blue Light Network on Dark Background
By
–
5. Illustration numérique de n’importe quel sujet Prompt pour ChatGPT-4o :
"Une illustration numérique de [SUJET], représentée à l’aide d’un réseau de lignes bleues lumineuses, nettes et épurées, qui soulignent son anatomie. L’image est placée sur un fond sombre, mettant en