I vote ComfyUI + Blender. Sounds like fun!
GENERATIVE AI
-
RL Verification Rewards vs SFT Synthetic Data Training
By
–
this isn't right. RL with verification-based-rewards is a different learning signal than generating synthetic data via language modeling and then using SFT, which is what you're describing
-
Practical AI Tools and Applications Coming Soon
By
–
Will be a mix, but lots of focus on stuff people can actually use right away!
-
Google IO 2026: Major AI Progress Announcements Coming
By
–
So much hard work by so many teams to make Google IO possible and exciting, can’t wait to be able to share all the progress : )
-
Reinforcement Fine-Tuning: o4-mini Learns from Minimal Examples
By
–
Reinforcement fine-tuning still feels magical to me.
— Romain Huet (@romainhuet) 12 mai 2025
Compared to regular fine-tuning, RFT on o4-mini learns from just a handful of solid examples to improve model performance for your use case, even in complex domains. https://t.co/gqyrKW3DqiReinforcement fine-tuning still feels magical to me. Compared to regular fine-tuning, RFT on o4-mini learns from just a handful of solid examples to improve model performance for your use case, even in complex domains.
-
ChatGPT Enhances Research Report Exports with PDF Functionality
By
–
Chatgpt devient de plus en plus utile, chaque jour.
— VISION IA (@vision_ia) 12 mai 2025
Traduction : Vous pouvez désormais exporter vos rapports de recherche approfondie au format PDF bien mis en page — avec tableaux, images, citations cliquables et sources.
Il suffit de cliquer sur l’icône de partage, puis de… https://t.co/0w33v9uf2WChatgpt devient de plus en plus utile, chaque jour. Traduction : Vous pouvez désormais exporter vos rapports de recherche approfondie au format PDF bien mis en page — avec tableaux, images, citations cliquables et sources. Il suffit de cliquer sur l’icône de partage, puis de
-

Google Uses Generative AI Veo Transform Product Images 3D
By
–
To help replicate the intuitive nature of shopping on a screen, learn how we’re using the latest #GenerativeAI models (including Veo) to transform 2D product images into immersive 3D visualizations for Google Shopping — from as few as 3 product images →https://t.co/0v13OL3N5t pic.twitter.com/rs7nA42DAZ
— Google AI (@GoogleAI) 12 mai 2025To help replicate the intuitive nature of shopping on a screen, learn how we’re using the latest #GenerativeAI models (including Veo) to transform 2D product images into immersive 3D visualizations for Google Shopping — from as few as 3 product images →
https://
goo.gle/4ddjJGJ -
AI to Generate Functional Apps from Simple Drawings
By
–
Vous dessinez une appli dans Paint, vous la capturez, vous écrivez : « Code-moi ça. »
— VISION IA (@vision_ia) 12 mai 2025
Et l’IA vous recrée l’application… fonctionnelle. En 2 minutes. Pas de prompt complexe. Pas de ligne de code. Juste une image.
Bientôt, on génèrera des produits entiers. pic.twitter.com/wADRw4OCoGVous dessinez une appli dans Paint, vous la capturez, vous écrivez : « Code-moi ça. » Et l’IA vous recrée l’application… fonctionnelle. En 2 minutes. Pas de prompt complexe. Pas de ligne de code. Juste une image. Bientôt, on génèrera des produits entiers.
-
Larger LLMs Easier to Control Than Smaller Models
By
–
Lots of fear out there about AI getting out of control as models get bigger. I think it's actually the opposite. The bigger LLMs get, the easier they are to control. Just a few gens ago, they were much harder to steer, stylize, mold. Scale doesn't hurt control – it helps.
