Google still lists gemini-3.1-flash-live-preview as free in the Gemini API as of late March 2026 — but the reality for developers is considerably more nuanced than the billing page suggests. Rate limits are no longer displayed in a central public table; instead, they live inside individual AI Studio projects, making

Google has shipped a meaningful — and mostly unannounced — upgrade to Gemini Live, the real-time voice conversation layer built on Gemini Flash. Users and developers working with the Live API are reporting noticeably more natural conversational flow, faster response cadence, and fewer awkward pauses that have historically made voice AI feel

Google's Gemini lineup now includes both Gemini 3 Pro and Gemini 3.1 Pro, and while the version bump might look incremental, there are meaningful differences that developers need to understand before choosing one for a production build. The headline addition in 3.1 Pro is Google Maps

Switching AI assistants has always come with a hidden tax: you lose the conversational context, preferences, and memory you've built up over months of use. Google is betting that removing that friction could be its most powerful weapon against OpenAI. Enter "switching tools" — a new set

Google TV just got a serious AI upgrade. On March 28, 2026, Google rolled out three new Gemini-powered features to compatible devices in the U.S. and Canada — and the living room experience will never quite be the same. Instead of just streaming shows, your TV can now serve up

Google has shipped a meaningful upgrade to the Gemini API focused on something developers have been asking for: a cleaner way to combine Google's built-in tools with custom developer functions in a single workflow. Previously, mixing native capabilities like Search or code execution with your own custom tools

Google's March 2026 Gemini Drop is the most ecosystem-integrated update the Gemini app has seen yet. The headline feature is Personal Intelligence coming to free users — meaning the app can now pull context from Gmail, Google Drive, Photos, Maps, and YouTube to give responses that actually know who

Google launched Gemini 3.1 Flash Live on March 26, 2026, and it's a meaningful step forward for developers building real-time conversational and multimodal applications. Available now in AI Studio and the Gemini app, the model is purpose-built for low-latency audio and voice interactions — the kind where a

Google just made a bold move in the AI coding wars: Gemini Code Assist is now completely free for individual developers, no credit card required and no stripped-down trial tier to navigate. As of March 2026, any developer can install the full VS Code or JetBrains plugin and get started