@googleaidevs
AI for every developer. So what will you build?
🗣️💬 Say hello to Gemini 3.5 Transcribe, a speech-to-text model that actually understands your codebase. Watch how this latest model filters out spoken hesitation while grounding technical terms, file names, and code variables precisely against the active context while building in @Antigravity. By using visual biasing, the model brings deep, screen-aware context to complex developer workflows.
Learn more to start building: blog.google/innovation-and…
From a single prompt to a fully animated parallax landing page. 🤯 See how Gemini 3.7 Flash, combined with Nano Banana and Omni, builds interactive websites in one shot. Under the hood, 3.7 Flash calls the right tools to generate the copy, image references, and render parallax-ready Omni videos.
Build your own: github.com/nroadley/Creat…
Turn the web into an interactive visual encyclopedia with Nano Banana and Gemini Omni. 🌐✨ We built a Chrome Extension in @Antigravity with Gemini 3.7 Flash that lets you highlight any text to quickly generate rich, visual definitions while you browse for deeper learning. Check out the demo ↓
We fed Gemini 3.7 Flash hundreds of PDFs from a Victorian book of botanical illustrations and tasked it with extracting and classifying these historical plants using the modern APG IV classification system. 🌳 💻 Watch how it brings its understanding to life in an interactive visualization built in @Antigravity:
You’ve planned your idea and built your app. Now it’s time to grow! 📈 Learn how you can market your product and deploy creative campaigns using tools like @FlowbyGoogle and @PomelliByGoogle for the Build with Gemini @XPrize Hackathon, live through August 17.
Get more details and submit before August 17: geminixprize.com
One spec. Five native apps. Zero boilerplate. Watch how Gemini 3.7 Flash in @antigravity takes a single architecture specification and generates native, production-ready code across 5 mobile frameworks: Flutter, SwiftUI, Jetpack Compose, React Native, and NativeScript.
Learn more: blog.google/innovation-and…
As our most intelligent workhorse model yet for coding and agents, Gemini 3.7 Flash brings higher instruction adherence, first-pass code accuracy, and high-quality execution for agentic tasks. See how the model handles building a complex 3D web game with 3.7 Flash in @antigravity: 1️⃣ 3D Engine Logic: Generates Three.js boilerplate, camera movement, and collision detection. 2️⃣ Asset Orchestration: Creates PBR-textured materials, custom sprite sheets with Nano Banana, and particle effects. 3️⃣ Procedural Audio: Integrates procedural sound effects using custom audio skills
Learn more: blog.google/innovation-and…
Real-time AI at 100 mph means moving your entire stack to the edge. Our @GoogleDevExpert community hit the track to build an offline AI racing coach that processes live car telemetry instantly, without relying on the cloud. Here’s what’s happening under the hood 🏎️ ↓
Step 1: Getting the data. You can't rely on cell service on a race track. The team wired a Pixel 10 directly into the car's network via a custom USB rig. Python scripts pulled in a 10 Hz telemetry stream from hundreds of physical sensors straight to the phone.
Use these @Antigravity commands to navigate the CLI, manage background tasks, configure utilities, and run research subagents more efficiently. 🚀 x.com/i/article/2085…
After planning your idea, it’s time to build. 🏗️ This video shows you how you can use the Google ecosystem to construct and deploy AI-native apps seamlessly for the Build with Gemini @XPrize Hackathon, live through August 17.
Get more details: geminixprize.com
After planning your idea, it’s time to build. 🏗️ This video shows you how you can use the Google ecosystem to construct and deploy AI-native apps seamlessly for the Build with Gemini @XPrize Hackathon, live through August 17.
Get more details: geminixprize.com
Gemini Robotics ER 2 is our most capable embodied reasoning model designed for physical AI 🤖 Built as a high-level brain for robotics, the model connects directly to the Gemini Live API. It processes continuous video streams to track progress, call tools, search the web, and command an action model in real time. Upgrades for developers include: + Enhanced progress tracking + Real-time mid-execution failure detection + Multi-robot collaboration + Advanced safety instruction following
Build smarter, safer physical agents today using the Gemini API or @GoogleAIStudio ↓ blog.google/innovation-and…
We wanted to see how Gemini 3.5 Flash-Lite handles massive, repetitive visual tasks. This demo runs the model across 1M+ catalog images, extracting raw features into clean, structured data with the low latency and token efficiency required for large-scale workflows.
👾Watch a multi-agent system built with Gemini 3.6 Flash iterate on playable game design in real time. It uses Gemini 3.5 Flash-Lite to design, construct, and verify puzzle challenges based on live player actions balancing speed and reasoning in a fast-paced environment.
Developer guide: ai.google.dev/gemini-api/doc…
Building for real work. See how our partners are integrating the latest Gemini 3.6 Flash and 3.5 Flash-Lite models into the tools you use every day. ↓
Building production AI agents requires scalable intelligence with token efficiency and lower latency. Today, we’re releasing Gemini 3.6 Flash and Gemini 3.5 Flash-Lite built to scale agentic workflows. Read the 🧵to learn more ↓
Gemini 3.6 Flash directly addresses your feedback, reducing token usage and overall cost compared to 3.5 Flash, requiring fewer reasoning steps and tool calls to accomplish multi-step workflows, making agents more cost-effective to build and run. Check out 3.6 Flash’s speed and token efficiency in this app that uses a Managed agent to source and analyze financial research. With 3.6 Flash, the agent accomplishes the same task faster with substantially fewer tokens used (when compared to 3.5 Flash) while searching for, analyzing, and creating a UI to report on the data gathered.
The Build with Gemini @xprize Hackathon is live through August 17, 2026. Dream big, build a startup that solves real-world problems, and compete for $2M in prizes. This video shows you how to use different Google products and tools, like Gemini Deep Research, Gemini Spark, and @GoogleAIStudio, to bring your project from idea to launch 💡
The Build with Gemini @xprize Hackathon ends August 17, 2026. Get more details: geminixprize.com
To help you deploy reliable, cost-controlled, and automated agents, we’re launching a few new updates to Managed Agents in Gemini API: - Free tier availability: Start building without active billing under free rate limits in Gemini API - Budget guardrails: Use max_total_tokens inside agent_config to prevent runaway task costs - Scheduled execution triggers: Automate recurring agent tasks with built-in cron schedules Read the article and start building ⬇️
Rather than translating text step-by-step, Gemini 3.5 Live Translate processes raw audio in real time. This eliminates lag and ensures every nuance of the speaker's natural voice is preserved. See how developers are building impressive multilingual apps with it ⬇️ x.com/googleaidevs/s…
Bring your entire repo into the context window and start building. @GoogleAIStudio now supports direct GitHub imports.
🧪From lab research to developer reality. Learn how we turn advanced AI research into practical tools you can actually build with in this talk from Benoit Schillings, VP of Technology @GoogleDeepMind at @aiDotEngineer.
We’re expanding managed agents in the Gemini API with added support for: 1. Long-running background execution: Run interactions asynchronously on the server without needing to hold an open HTTP connection. 2. Remote MCP server: Connect managed agents directly to private databases, internal and external APIs with no proxy middleware required. 3. Custom function calling: Add custom tools seamlessly with built-in sandbox tools. 4. Network credential refresh: Rotate API keys mid-session without losing your sandbox's filesystem state. Start building autonomous agents today ↓ x.com/GoogleAIStudio…
Check out our Quickstart guide: ai.google.dev/gemini-api/doc…
As we build tools that scale alongside rapid AI advancements, platform design must evolve to let the models do their best work. @kevinhou22 on the evolution of @Antigravity at @aiDotEngineer.
We’re launching Nano Banana 2 Lite, our fastest most cost-efficient Gemini Image model yet, optimized for high-throughput developer pipelines. Alongside it, Gemini Omni Flash, our video generation and editing model is now available to developers. Both models are accessible today via @GoogleAIStudio and the Gemini API to help you scale your workflows 👇
Nano Banana 2 Lite is now available for developers to build high-velocity image pipelines. The model generates text-to-image outputs in 4 seconds at a cost of $0.034 per 1K image, enabling quick drafting and prototyping at a low cost for developers.
Host a sponsored Gemma hackathon for your local community. 🌐 x.com/googlegemma/st…
When we hit 2 million apps, we have to build the real statue, right? x.com/GoogleAIStudio…
Try it out: aistudio.google.com/apps?features=…
The Computer Use tool is now available for Gemini 3.5 Flash 🛠️ Build agents that see and take action across browser, mobile, and desktop environments to seamlessly handle long-horizon tasks. Here’s what’s new: - Built-in support for mobile and desktop operating systems - Intent arguments for all function calls - Customizable client-side functions to easily support human-in-the-loop (HITL) takeover - Prompt injection detection and configurable action-level safety policies to fit your app’s needs Learn how you can use 3.5 Flash to automate QA testing, business processes, and more in the blog: t.co/WwkqWt9Rcu
Watch the tool in action as it completes a doc audit for mobile accessibility issues:
The Gemini Interactions API is your unified interface for text, multimodal inputs, tool use, and managed agents all in a single endpoint. Go from prompt to production faster than ever. Read the developer guide ⬇️ x.com/GoogleAIStudio…
🗣️We built Managed Agents in the Gemini API to handle the plumbing so you can focus on building smart, autonomous agents. With just a single prompt, see how the agent researches a topic and builds a radio show from scratch. Here’s how it works: - [Zero infrastructure setup] The API automatically spins up a secure, temporary Linux sandbox. - [Autonomous loops] The agent independently plans, self-corrects, and executes its own code. - [Multi-step logic] With instructions and tools defined via industry-standard agents.md and skills.md files, the agent uses its Gemini-optimized understanding to draft and produce a dynamic, back-and-forth audio dialogue, generate music, create album art and more, all in a single API call.
Try it for yourself in @GoogleAIStudio: aistudio.google.com/apps/bundled/a…
The Interactions API is now GA. Featuring a stable schema and new capabilities via a single /interactions endpoint, it is now our standard interface for both model inference and autonomous agents. Check out the key updates: — Managed Agents: Spin up a secure, remote Linux sandbox for code execution and browsing via a single API call, using the default antigravity-preview-05-2026 agent or your own custom instructions — Simplified Schema: Every action is now represented as its own step (from user_input to model_output, etc.) in a simplified, linear flow — Background Execution: Set background=True to run models and agents asynchronously, letting you poll for results when ready — Expanded Tool Ecosystem: Native support for Computer Use, File Search, and Google Maps, alongside new agent-building tools like the Gemini API Docs MCP server and gemini-interactions-api Skill — Media Generation: Image generation via Google Search-grounded Nano Banana 2, music with Lyria 3, and expressive multi-speaker TTS Read the blog to learn more: t.co/ZzfpUGzAvm
Check out this quick overview of the @Antigravity ecosystem 🚀
Watch the full episode to get the breakdown on Antigravity 2.0 and the IDE, CLI, and SDK ⤵️ youtube.com/watch?v=04IqH3…
How does Gemma 4 12B handle audio and vision without encoders? @MaartenGr maps out the architecture in this step-by-step visual guide. ⤵️ newsletter.maartengrootendorst.com/p/a-visual-gui…
Learn how to vibe code in 5 days! Build scalable agent systems using natural language and complete a hands-on capstone project in this @Kaggle course hosted by our researchers and engineers.
Register here: kaggle.com/competitions/5…
Add near real-time voice translation to your apps with Gemini 3.5 Live Translate via the Gemini Live API. 🎙️ Watch how the model handles live broadcast ingestion and translation with continuous speech-to-speech streaming (S2ST) and synced transcripts, letting users tune into
Try building in @GoogleAIStudio and show us what you make. aistudio.google.com/live?model=gem…
DiffusionGemma, our experimental open model released under an Apache 2.0 license, explores text diffusion, an exceptionally fast approach to text generation. Here’s how DiffusionGemma accelerates development: + Faster token output: By shifting the bottleneck from memory
Get more details in the blog: goo.gle/3Sw3MoC
Our latest audio model, Gemini 3.5 Live Translate, takes real-time speech translation to the next level for developers by delivering low-latency translation across 70+ languages. By processing speech as it streams in near real time, the model enables devs to build low-latency
Here’s how you can start building today: — @GoogleAIStudio: Test model capabilities in the live playground at ai.studio/live — Gemini Live API & 3P integrations: Get started with built-in ecosystem support from @AgoraIO, @livekit, @pipecat_ai, @swmansion,
New @GoogleGemma 4 QAT (Quantization-Aware Training) checkpoints are here, so you can run models locally on consumer GPUs and mobile devices with minimal quality loss. What’s new: 🔹 GGUF (Q4_0): Checkpoints: Max local performance across all sizes and drafter models 🔹 Custom
Download the weights on @HuggingFace today. huggingface.co/collections/go…
Play our new open-weights music model, @GoogleMagenta RealTime 2, using a MIDI keyboard, live text prompts, and even hand gestures ✌️ x.com/GoogleMagenta/…
Join @GoogleDeepmind and @HeyGen on June 11th! Our LA event for builders working at the intersection of AI agents, creative tooling, and multimodal apps is now open for registration 👇 x.com/HeyGen/status/…
We’re launching Gemma 4 12B: Our unified, encoder-free model that brings powerful multimodal intelligence straight to your laptop 🚀 The model bridges the gap between our mobile E4B model and larger 26B MoE models, packaging frontier-class reasoning and native audio into a
> Ecosystem: Compatible with llama.cpp, MLX, @LMStudio, vLLM, @ollama, @UnslothAI, and SGLang. > Download: Grab the weights on @Kaggle and @HuggingFace: huggingface.co/collections/go…
Building autonomous agents for scientific discovery? 🧬🤖 @GoogleDeepMind Science Skills is now available on GitHub. We've open-sourced this specialized toolkit to accelerate your agentic workflows with scientific grounding and higher token efficiency. Download now ↓
Check out how you can use Science Skills ↓ antigravity.google/use-cases/scie…
From Google I/O to building real world solutions in one weekend. Always inspired by the energy of this community! x.com/cerebral_valle…
From clutter to clarity. Watch parallel subagents in @Antigravity sort and rename hundreds of marketing assets, eliminating manual file management.
ICYMI: Nano Banana Pro [gemini-3-pro-image] and Nano Banana 2 [gemini-3.1-flash-image] are now GA and ready for production via the Gemini API. Check out these great community examples to see the capabilities of both models in action 🧵
Advanced world knowledge: x.com/BoWang87/statu…
Good implementation of Managed Agents in the Gemini API by our friends at @Eigent_AI! Root causing CI failures at lightning speed. ⚡ x.com/Eigent_AI/stat…
The team at @llama_index built an awesome template using LlamaParse and the new Managed Agents in the Gemini API. See how they built an agent that can tackle unstructured documents. 📄↓ x.com/llama_index/st…
We love this use of Managed Agents in the Gemini API by the team at @wandb. Silent bugs don't stand a chance. 🐛⬇️ x.com/wandb/status/2…
Save countless hours of manual migration work. Watch messy legacy code autonomously modernize into Next.js in this example from @Antigravity, powered by Gemini 3.5 Flash.
Start building today: ai.google.dev/gemini-api/doc…
