Google I/O, Claude 4, Veo 3, open-source image editors, open-source agents, AI traffic cams, AI improves batteries
Welcome to the AI Search newsletter. Here are the top highlights in AI this week.
Claude 4 is Anthropic’s newest AI model for advanced coding and reasoning. It comes in two versions—Opus 4 for top-tier coding and Sonnet 4 for faster, precise responses—with features like tool use, improved memory, and seamless code integration. Both are available through Anthropic’s API and major cloud platforms for developers and teams. Read more
Veo 3 is Google’s latest AI that generates videos with soundtracks, sound effects, and even dialogue. It’s a big upgrade from Veo 2, producing higher-quality footage and letting users create videos by describing scenes or uploading images. Veo 3 is available through Google’s Gemini chatbot for Ultra plan subscribers. Read more
Google Jules is a fully autonomous AI coding agent that can read, understand, and work on your code by itself. It automates tasks like writing tests, fixing bugs, and updating dependencies, working directly with your GitHub projects without manual setup. Jules is now available to all developers where Gemini is supported. Read more
Do you prefer to watch instead of read? Check out this video covering all the highlights in AI this week:
Scientists used AI and X-ray analysis to understand how zinc-ion batteries work. They found that high concentrations of a certain salt in the battery's electrolyte help stabilize water molecules and improve the battery's performance. This discovery could lead to more efficient and sustainable energy storage solutions. Read more
AI-powered traffic cameras have been shown to significantly reduce traffic accidents, injuries, and property losses by promoting safer driving behaviors. These cameras use artificial intelligence and continuous video recording to detect a wider range of violations and deter unsafe driving, without shifting risk to nearby locations. This technology has the potential to improve road safety and save lives. Read more
We’re partnering with Dell to give away a Dell Precision 5690 Workstation with a RTX 5000 Ada. This is a powerful yet portable laptop, with a built-in NPU that’s optimized for AI. Only available to USA or Canada residents. Enter for FREE here
Bytedance BAGEL is a 7B parameter AI model that can generate, edit, and understand images all in one. It’s a unified multimodal model, so you don’t need separate tools for different tasks—just use BAGEL for everything from text-to-image to image editing and analysis. Read more
Devtral is Mistral’s new platform for building and deploying AI agents and workflows. It’s designed to help developers create, manage, and scale agentic AI systems with better tools and integrations, making it easier to automate complex tasks. Read more
Magentic-UI is an open-source research prototype for a human-centered, multi-agent AI interface. It lets users browse the web, generate and execute code, and analyze files, all powered by a system of collaborating AI agents for more advanced automation. Read more
MedGemma is Google’s new medical AI model built for analyzing medical images and clinical text. Based on the Gemma3 architecture, it comes in two sizes: 4B for image analysis (like X-rays) and 27B for processing clinical notes, supporting doctors in diagnosis and decision-making. Developers can use it locally or via Google Cloud for healthcare applications. Read more
With Monica, you can use the top AI models, image generators, and video generators, all in one integrated platform. Use code AISEARCH10 to get 25% OFF 'Unlimited Annual Plan' within 24h of registration, or enjoy 10% OFF. Try it for free today!
Gemini Diffusion is Google’s experimental AI for generating and editing text much faster and more coherently. It refines whole blocks of text step-by-step, making it great for editing, coding, and math tasks, and can fix errors as it goes. The model is notably quicker than previous versions, generating 1479 tokens per second. Read more
Google Stitch is an AI tool for quickly generating user interfaces and frontend code for apps. Just describe your idea or upload a sketch, and Stitch creates ready-to-edit HTML/CSS or Figma designs, making UI development much faster. It’s free with monthly usage limits and supports exports to Figma or code editors. Read more