Google AI Agents Hackathon

Build powerful AI Agents with the latest Google AI tech

⏱️ You’ve got 6 days — design, build, and showcase intelligent AI Agents in a fast-paced global hackathon.

🌟 Build with Google’s AI stack, including Gemini models, Google AI Studio, and emerging agent frameworks with guidance, docs, and hands-on experimentation.

🤝 Work solo or with a team to bring autonomous, multimodal, or research-driven AI Agents to life.

💰 $50,000 PRIZE POOL for the best AI-powered solutions.

🧑‍💻 Sign up and start building with Google’s latest AI tools.

About

Gemini is Google’s family of next-generation multimodal AI models designed to power intelligent, agent-driven applications. Gemini can reason across text, images, code, video, and audio, and supports Function Calling, enabling agents to interact with tools, APIs, and workflows in real time.

Google AI Studio is a browser-based environment where developers can easily explore and prototype with Gemini models. It allows you to test prompts, refine agent behavior, and experiment across modalities before integrating your agents into real-world applications through the Gemini API.

This challenge also highlights Google’s Live API, enabling real-time, low-latency multimodal interactions with Gemini models. Participants can build applications that stream audio, video, or screen input over WebSockets, allowing for live conversations, voice agents, and interactive experiences powered by Gemini.

In this hackathon, builders will design, prototype, and ship AI Agents that think, decide, and act — showcasing what’s possible with Google’s latest AI infrastructure.

Challenge

Build powerful AI Agents using Google’s latest AI technology.

Your mission is to design and implement an AI Agent that demonstrates strong reasoning, autonomy, and real-world usefulness using Gemini models and Google AI tooling.

Your agent should go beyond simple chat or prompt-response behavior and showcase at least one of the following:

What your agent should demonstrate

🧠 Intelligent Reasoning
Use Gemini’s reasoning capabilities to analyze inputs, synthesize information, and make decisions based on context, goals, or constraints.

🔄 Agentic Workflows
Design workflows where the agent plans steps, calls tools or functions, manages state, or coordinates multi-step tasks over time.

🌍 Real-World Usefulness
Solve a real problem — for individuals, teams, or organizations — such as research, productivity, automation, data analysis, content creation, or operational support.

🎙️ Live & Real-Time AI Experiences
Use the Gemini Live API to build real-time voice, audio, video, or screen-streaming applications with low latency, interruptible responses, and live function calling.

🧩 Multimodal Intelligence (Optional but encouraged)
Leverage Gemini’s ability to work with images, documents, code, audio, or video to unlock richer interactions and insights.

Technology

To build your project, you will use Gemini models and Google AI tooling to power the intelligence and behavior of your agent.

Participants should select the Gemini model that best aligns with the intelligence, speed, and complexity required by their AI Agent.

Gemini 3 Flash is optimized for speed and responsiveness. It’s ideal for agents that need to operate in real time, handle frequent interactions, or power fast user-facing experiences such as chat assistants, workflow automations, and interactive tools. If your agent prioritizes low latency and high throughput, Gemini 3 Flash is a strong starting point.

Gemini 3 Pro is built for advanced reasoning and deeper cognitive tasks. It’s well suited for agents that need to analyze complex inputs, reason across long contexts, plan multi-step actions, or synthesize information from multiple sources. Teams building research agents, analytical assistants, or decision-making systems will benefit most from Gemini 3 Pro’s enhanced reasoning capabilities.

Access to Technology:

Participants can explore Gemini through two main options - both include free access:

🌐 Google AI Studio: A free, browser-based IDE for experimenting with Gemini. You can write prompts, test outputs, tweak parameters, and prototype without needing to set up billing or any API keys. Perfect for prototyping and rapid iteration during the hackathon.

💻 Gemini API: For developers ready to integrate Gemini directly into their apps. The free tier includes a monthly allowance of tokens for text and multimodal inputs/outputs, letting you test and build with Gemini’s core models at no cost. Paid usage only applies for higher-volume or production-level apps.
🎙️ Live API: The Gemini Live API enables real-time, low-latency multimodal interactions with Gemini models using WebSockets. It supports streaming audio, video, and screen input for building live voice agents, interactive assistants, and real-time AI experiences.

🔓 Google Cloud Credits:
New Google Cloud accounts receive $300 in free Google Cloud credits, valid for 90 days.
These credits can be used across eligible Google Cloud services, including Gemini API usage, subject to Google Cloud’s standard terms and regional availability.

👉 Learn more about Google Cloud’s free tier

Note: After the free credits are used or expire, continued usage of billable services requires enabling billing on your Google Cloud account. Charges only apply if you choose to continue using paid services.

🚀 How to Get Started

1. Visit the Gemini docs
2. Follow the Quickstart Guide to explore Gemini in AI Studio
3. Experiment with text, code, images, video, or audio prompts
4. Build and test your multimodal agents or workflows

📘 Resources:

• Gemini API models
• Gemini 3 Developer Guide

• Gemini API pricing & models docs

• Gemini API billing FAQ & limits

• Live API reference (Vertex AI)

• Google AI Studio official overview

• Google AI Studio quickstart
• Antigravity

Prizes

Winner

☁️ $25,000 in GCP credits

2nd Place

☁️ $15,000 in GCP credits

3rd Place

☁️ $10,000 in GCP credits

Please be aware that participation in lablab.ai hackathons is voluntary and subject to the specific rules and conditions outlined for each event. While we strive to provide a fair and rewarding experience for all participants, we cannot guarantee entry or success in securing prizes. All prizes and opportunities are subject to availability and participants must meet the established criteria to qualify. lablab.ai is not responsible nor held liable for any commitments or offers - including programs, giveaways, or prizes - made by third-party sponsors or partners of the hackathon. The availability and delivery of third-party prizes or benefits are solely at the discretion of the respective sponsors and are not guaranteed by lablab.ai. We reserve the right to make amendments, modifications, or cancellations to any part of the hackathon (including full cancellation of the event itself), including prizes and terms of participation, at our sole discretion. Whenever possible, we will endeavor to notify participants of significant changes within a reasonable timeframe. All submissions by participants must be original work, open source, and compliant with the MIT License unless specified otherwise. By participating, you agree to these terms, the overall event terms, and the lablab.ai general terms and are encouraged to review all event-related information carefully. Prize Distribution: lablab.ai and its partners reserve the right to distribute all prizes within a 90-day period following the end date of the hackathon. This period allows for the proper processing and verification needed to award prizes appropriately.

What to submit?

📋 Basic Information

• Project Title

• Short Description

• Long Description

• Technology & Category Tags

📸 Cover Image and Presentation

• Cover Image

• Video Presentation

• Slide Presentation

💻 App Hosting & Code Repository

• Public GitHub Repository

• Demo Application Platform

• Application URL

For further details and guidance, please visit Submission Guidelines

lablab.ai

Topic: Hackathon Submissions Process

▶️ Watch Video

Judging Criteria

Application of Technology

How effectively the chosen model(s) are integrated into the solution.

Presentation

The clarity and effectiveness of the project presentation.

Business Value

The impact and practical value, considering how well it fits into business areas.

Originality

The uniqueness & creativity of the solution, highlighting approaches and ability to demonstrate behaviors.

Hackathon Details

Join lablab.ai hackathon and innovate using the latest models in the market. Discover all the relevant details below.

🗓️ Where and when

The start date of the hackathon is mentioned according to the date specified on the hackathon page, cover and schedule. The hackathon will take place on the lablab.ai platform and lablab.ai Discord server.

🦸🏼‍♂️ Who can join?

Everyone is welcome to participate, regardless of previous AI or coding experience! We encourage anyone with a passion for AI or an interest in exploring how it can be used in their field to join.

😅 What about teams?

If you don't have a team, don't worry! You can connect with other participants from all over the world on our dashboard or Discord server. We also recommend checking out our Discord server to find teammates and bounce around ideas. You can join the server here.

🛠️ How to participate

The hackathon will take place online on lablab.ai platform and lablab.ai Discord Server. Please register for both in order to participate. To participate click the "Sign up" button at the bottom of the page and read our Hackathon Guidelines and Getting Started Guide.