We’ve been making relentless progress over the last few months across Gemini models and App: 2.5 Pro, Deep Research, Canvas, multimodal image gen, video understanding and more. Super excited to be working with the brilliant @JoshWoodward to take the @GeminiApp to the next level!
MULTIMODAL AI
-
AI Video Re-style Controls: Depth, Edges, Pose Motion Capture
By
–
we offer 4 main controls for Video Re-style.
— KREA AI (@krea_ai) 3 avril 2025
with "Depth", the AI will keep the motion that captures from a depth map on the original video.
"Edges" will capture the motion from the edges on the original video – great for close-ups or faces talking.
"Pose" will only capture… pic.twitter.com/rN0oH5yzn0we offer 4 main controls for Re-style. with "Depth", the AI will keep the motion that captures from a depth map on the original video. "Edges" will capture the motion from the edges on the original video – great for close-ups or faces talking. "Pose" will only capture
-
Language Supervision Unnecessary for Multimodal Visual Representations
By
–
New paper from FAIR+NYU:
Q: Is language supervision required to learn effective visual representations for multimodal tasks? A: No. -
Gemini 2.5 Pro Launches on Poe as World’s Most Powerful Model
By
–
Now on Poe: Gemini 2.5 Pro! This is the most powerful model in the world today according to a few benchmarks, and it is #1 ranked on LMArena by a large margin. It supports text, image, video, and audio input, and has a 1M token context window. (1/2)
-
Gen-4 Brings Objects, Locations and Characters to Life
By
–
Bring any object, location or character to life with Gen-4. pic.twitter.com/tDReGCT9r7
— Runway (@runwayml) 3 avril 2025Bring any object, location or character to life with Gen-4.
-
AI Generated Magazine Layouts in 31 Seconds: Industry Disruption
By
–
Magazine setters are cooked, AI made this in 31 seconds. Now what?
-
Streaming video generation available in a few weeks
By
–
As soon as we have streaming video generation. In a few weeks :-).
-
Comparative Performance Test of ChatGPT and Grok 3 Image Generation
By
–
ChatGPT's new image generator is wild. But can it beat Grok 3? I tested both with 10 creative prompts to find out which one is better. And the results shocked me (The final prompt serves as the ultimate test for all these image generators)
-
Real-time voice and video AI conversation project
By
–
Yesterday, I spoke with a company where two people explained to me that they had done an "impressive" project with two outsiders last year. They said they had coded a real-time, two-way conversation with a voice AI (custom voice model) and video (y
-
Using ChatGPT to generate an infographic on tariffs
By
–
"ChatGPT, create an infographic image explaining to me how tariffs work like I'm five"