Google launched Gemini Robotics ER 2, an embodied reasoning model that helps robots understand surroundings, communicate, and complete complex physical tasks in real time, available through the Gemini…
Google launched Gemini Robotics ER 2 on July 30, making the embodied-reasoning model publicly available through the Gemini API and Google AI Studio. Google reports 57.4% accuracy on progress classific…
Google DeepMind launched Gemini Robotics ER 2, its most capable embodied reasoning model for robotics, enabling robots to understand video feeds, orchestrate multi-step tasks, and collaborate in share…
Google DeepMind has introduced Gemini Robotics ER 2, an embodied-reasoning model for video understanding, task orchestration, and multi-robot collaboration, but it is not yet released and has no annou…
Anthropic's Claude Opus 5 supports a 1M token context window and up to 128k output tokens, but developers must design a long-output architecture to handle reliability issues like timeouts, stream drop…
Google has added environment hooks to managed agents in its Gemini API, letting developers run custom code before and after every tool call and cancel calls outright, with Gemini 3.6 Flash becoming th…
Google has updated its Gemini API managed agents with Gemini 3.6 Flash as the default model, adding environment hooks, token budget controls, scheduled execution, sandbox management, and free tier acc…
Google's Gemini API Managed Agents now default to Gemini 3.6 Flash, introduce environment hooks for custom pre- and post-tool execution scripts, and offer free tier access, enabling a single API call …
Google shipped four production-readiness upgrades to Gemini API Managed Agents on July 7, including background execution that allows agents to run asynchronously server-side without a persistent HTTP …
Google released its first ATLAS study on July 23, analyzing about 15 million de-identified interactions across the Gemini app, AI Mode and Gemini API, finding AI activity across 68% of occupations rep…
Developers must adopt a reasoning budget policy for LLMs to avoid wasting latency and cost, according to a new guide analyzing Anthropic's Claude Opus 5, OpenAI's reasoning models, and Google's Gemini…
Google researchers found that Americans ask Gemini and AI Mode about government services, health, money, and legal issues far more often than they spend time on those activities, while eating, dressin…
Google launched the first iteration of the AI & Economy ATLAS (Activity, Task, Landscape, and Adoption Study), a large-scale, de-identified study of how people use Google's AI products, built from 15 …
Google launched Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber, targeting developers and enterprise customers building AI agents. Gemini 3.6 Flash achieves a 17% reduction in toke…
Google shipped three new Gemini models — 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber — on April 9, 2025, competing on cost per task rather than frontier benchmarks. Gemini 3.6 Flash costs $1.50 per…
Google announced three new AI models today, including Gemini 3.6 Flash and its first cybersecurity-focused model, Gemini 3.5 Flash Cyber, while teasing upcoming Gemini 3.5 Pro and Gemini 4. The new Ge…
A developer built a free, serverless AI bot for Bluesky using GitHub Actions and Google's Gemini API to automate posting and engagement without manual effort or infrastructure costs. The bot runs on a…
Google Cloud announced that Parallel Web Systems is now a natively integrated web grounding provider on the Gemini Enterprise Agent Platform, enabling developers to ground Gemini models in real-time w…
A developer replaced a hardcoded NHL stats screen with an AI-powered assistant using Google's Gemini API, enabling users to ask natural-language questions like "Who are the top scorers this season?" i…
Google launched Nano Banana 2 Lite, an entry-level image generation model that produces images in four seconds at roughly $0.034 per image, about half the cost and 2.7 times faster than Nano Banana 2.…