cd /news/generative-ai/google-showcases-player-generated-ga… · home topics generative-ai article
[ARTICLE · art-31258] src=cryptobriefing.com ↗ pub= topic=generative-ai verified=true sentiment=↑ positive

Google showcases player-generated game built with Gemini API at Google I/O

Google debuted Infinite Scaler at Google I/O 2026, a game where attendees generated 3D levels in real time using text prompts and reference images via the Gemini API. The game showcased Gemini's speed at 1,500 tokens per second and integrated tools like Lyria 3 for adaptive soundtracks, highlighting Google's AI capabilities for creative applications.

read2 min views1 publishedJun 17, 2026

Infinite Scaler let attendees create 3D game levels in real time using text prompts and reference images, no traditional asset pipeline required

Google turned its annual developer conference into a live gaming experiment. At Google I/O 2026, held May 19-20 at Shoreline Amphitheatre, the company debuted Infinite Scaler, a game where players generated entire 3D levels on the fly using nothing but text prompts and reference images submitted through their phones.

The game ran on Google’s Gemini API, the same foundation model powering a growing list of the company’s AI products. Attendees were the content creators, feeding prompts into the system and watching as 2D sprites were converted into playable 3D WebGL environments in near real time.

How Infinite Scaler actually works #

The core loop is deceptively simple. A player submits a text prompt or uploads a reference image from their mobile device. The Gemini API processes that input, generates visual assets, and constructs a game level around them. Players then compete inside the environments they collectively built.

The technical stack behind this involved several Google tools working in concert. Google AI Studio served as the development backbone, while the Gemini API and Canvas handled the generative heavy lifting for visuals and level design. Nano Banana, a tool for image and map generation, created the visual assets. Google Antigravity enhanced what the team described as “player agency,” essentially giving users more meaningful control over the game world they were shaping.

Then there’s the audio layer. Lyria 3, Google’s music generation model, produced real-time soundtracks that adapted to the gameplay. So the levels weren’t just visually generated on the spot. They were scored on the spot too.

The speed was the real flex. During live demos, Gemini Flash models achieved game generation rates of approximately 1,500 tokens per second. For context, that’s fast enough to make the creation process feel nearly instantaneous to a player standing in a crowd at an outdoor amphitheatre, tapping on their phone.

What this means for investors #

The crypto angle here is, frankly, nonexistent. Google I/O 2026 featured no discussion of cryptocurrency tokens, blockchain integration, or decentralized gaming. This was a pure developer tools play, focused on showing what Gemini can do when pointed at creative applications.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our

Editorial Policy.

── more in #generative-ai 4 stories · sorted by recency
── more on @google 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/google-showcases-pla…] indexed:0 read:2min 2026-06-17 ·