Ahead of I/O, we’re releasing an updated Gemini 2.5 Pro! It’s now #1 on WebDevArena leaderboard, breaking the 1400 ELO barrier! Our most advanced coding model yet, with stronger performance on code transformation & editing. Excited to build drastic agents on top of this!
GENERATIVE AI
-
Model Builds Retro Windows 95-Style Web App from Simple Prompt
By
–
Here’s a glimpse of what it can do – check out this retro Windows 95-style web app the model can build and extend from a simple prompt. pic.twitter.com/YwBUnyHefs
— Jeff Dean (@JeffDean) 6 mai 2025Here’s a glimpse of what it can do – check out this retro Windows 95-style web app the model can build and extend from a simple prompt.
-
Google Unveils Gemini 2.5 Pro I/O Edition with Major Performance Gains
By
–
Today we’re sharing an early look at our latest Gemini update for I/O! Introducing the updated Gemini 2.5 Pro (I/O edition), which ranks #1 on WebDev Arena and surpasses our previous 2.5 Pro model by +147 Elo points.
-
Meta-Agentic: Self-Evolving Agent Factory Revolutionizes AGI Development
By
–
Meta-Agentic 👁️✨ — the master key 🗝️
— MONTREAL.AI (@Montreal_AI) 6 mai 2025
A self-evolving agent-factory that spins exponential design loops, harvests value across every vertical, and rewrites the rulebook as it learns.
Demo: https://t.co/YtFeoMMiT3
Own the future.#AGI #AGIALPHA #MetaAgentic pic.twitter.com/V2QC6sTkwBMeta-Agentic — the master key A self-evolving agent-factory that spins exponential design loops, harvests value across every vertical, and rewrites the rulebook as it learns. Demo: https://
github.com/MontrealAI/AGI
-Alpha-Agent-v0/tree/main/alpha_factory_v1/demos/meta_agentic_agi
… Own the future. #AGI #AGIALPHA #MetaAgentic -
Textbooks Have Zero Hallucination Rate Unlike AI
By
–
you're not wrong. the only difference is that textbooks have a 0% hallucination rate
-

Wan 2.1 Update: Start and End Frame Support
By
–
Wan 2.1 now works with start and end frames.
— KREA AI (@krea_ai) 6 mai 2025
try it now! pic.twitter.com/ImvlzI99jkWan 2.1 now works with start and end frames. try it now!
-
Runway Academy: 3D Object Positioning and AI-Powered Creative Workflow
By
–
A Runway Academy will be coming soon, but the process is simple: use your 3D program of choice to position your object how you’d like it to appear. Pre-comp that with a location plate and export as a 2D image. Combine that with your style reference image and location image using
-

Stable Diffusion 3.5 now native in ComfyUI platform
By
–
Stability AI @ComfyUI since day one. Together, we've laid the foundation for one of the most vibrant communities in generative AI. Now, it's easier to join in. Stable Diffusion 3.5 Large and Stable Image Ultra are now available in ComfyUI's new native API nodes, making
-

ContextGem: Open Source LLM Framework for Document Data Extraction
By
–
Extract structured data and insights from documents with just a few lines of Python code! ContextGem is an LLM framework that makes extracting structured data and insights from documents radically easier with minimal code. 100% Open Source
-
Gen-4 References: 3D Object Positioning with AI
By
–
With Gen-4 References you can use 3D models to quickly position objects exactly where you want them. With just three images: a location, a 3D asset and a style reference, you can generate highly refined outputs without any rigging, texturing or rendering. pic.twitter.com/NQtPXGhKJd
— Runway (@runwayml) 6 mai 2025With Gen-4 References you can use 3D models to quickly position objects exactly where you want them. With just three images: a location, a 3D asset and a style reference, you can generate highly refined outputs without any rigging, texturing or rendering.
