Looks like Google upgraded Custom Gems to be powered by Gemini 2.5 Pro. Some users are also reporting that they have Gemini 2.5 Flash instead. Will it make GEMs useful?
MULTIMODAL AI
-
Medical AI Assistant: Multi-Agent LangGraph Healthcare System
By
–
Medical AI Assistant A multi-agent system powered by LangGraph that combines medical diagnosis, imaging analysis, and voice interaction for comprehensive healthcare support. Check it out!
-
ReDi Framework Improves Image Generation via Joint Feature Synthesis
By
–
Boosting Generative Image Modeling via Joint Image-Feature Synthesis This paper introduces ReDi, a new framework that jointly models low-level VAE latents and high-level semantic features within the diffusion process, significantly improving image generation quality and training
-
GPT-Image Feedback Loop: Generating Photos Five Seconds Forward
By
–
Another gpt-image feedback loop: "Generate the same photo 5 seconds into the future."
— Gene Kogan (@genekogan) 26 avril 2025
Starting with distracted boyfriend meme. pic.twitter.com/xT1lOfibKyAnother gpt-image feedback loop: "Generate the same photo 5 seconds into the future." Starting with distracted boyfriend meme.
-
Gen-4 References Early Access Released for Gen:48 Participants
By
–
We have released early access to Gen-4 References to all Gen:48 participants. References allow you to create consistent worlds with consistent characters and locations. This early preview is already available to all teams participating in Gen:48 for free. Good luck with your… pic.twitter.com/ruGzZXf9VX
— Runway (@runwayml) 26 avril 2025We have released early access to Gen-4 References to all Gen:48 participants. References allow you to create consistent worlds with consistent characters and locations. This early preview is already available to all teams participating in Gen:48 for free. Good luck with your
-
ChatGPT o3 Identifies Paris Restaurant from Photo Without Metadata
By
–
CHATGPT O3 COMMENCE A TOUCHER L'AGI.
— VISION IA (@vision_ia) 26 avril 2025
Un simple menu photographié à Paris. Envoyé à ChatGPT. 1 min 22 plus tard : il identifie le restaurant… sans métadonnées.
Comment ?
→ Zooms successifs sur l’image
→ Analyse du moindre détail : nom de plat, couleurs, emblèmes
→… https://t.co/XBzd9P5xBM pic.twitter.com/t1gHc8Ck27CHATGPT O3 BEGINS TO TOUCH AGI. A simple menu photographed in Paris. Sent to ChatGPT. 1 min 22 later: it identifies the restaurant… without metadata. How?
→ Successive zooms on the image
→ Analysis of the slightest detail: dish name, colors, emblems
→ Contextual reasoning, -
Genie 2: Foundation World Model for Embodied Agent Training
By
–
Join Jack Parker-Holder & Philip Ball today at 5PM at the #ICLR2025 Google booth to discuss Genie 2, a foundation world model that can generate an endless variety of action-controllable, playable 3D environments for training & evaluating embodied agents. https://t.co/xPsrqa6j3s pic.twitter.com/5uioM3ejRy
— Google AI (@GoogleAI) 26 avril 2025Join Jack Parker-Holder & Philip Ball today at 5PM at the #ICLR2025 Google booth to discuss Genie 2, a foundation world model that can generate an endless variety of action-controllable, playable 3D environments for training & evaluating embodied agents. https://
deepmind.google/discover/blog/
genie-2-a-large-scale-foundation-world-model/
… -
GPT-4o Updated with Improved Intelligence and Personality
By
–
Annonce @sama mais sans plus de détails !! Trad : Nous avons mis à jour GPT-4o aujourd'hui! avec une amélioration de son intelligence et de sa personnalité.
-
Google Gecko: Interpretable Metric for Image Prompt Adherence
By
–
At #ICLR2025? Stop by the Google booth today at 12PM to learn about Gecko, an interpretable, automatic metric for evaluating the prompt adherence of images & videos. Given an image/video + description, the model determines what is missing from the prompt. https://
arxiv.org/abs/2404.16820
