Creators Developed Playable Worlds, VR Games, & Digital Pets for Tripo AI's First 3D World Hackathon

What does it mean to build a world?

For some creators, it means constructing a playable environment filled with characters, mechanics, and objectives. For others, it might be turning a child’s drawing into an interactive toy, giving historical figures conversational abilities, or placing a digital creature inside a piece of wearable jewelry.

Those different interpretations were on display during Tripothon S1, the first global world-building hackathon hosted by Tripo AI. More than 2,000 registered teams from 78 countries and regions participated across five categories: games, film and VFX, VR/XR/AR, applications, and physical design.

The competition ran from September 15 through October 20, 2026, with online submissions accompanied by in-person events in San Francisco, São Paulo, Seoul, Barcelona, and Shanghai. Participants worked from the theme “A Gift for ___,” challenging them to create a world for a person, place, or future that might not yet exist, utilizing Tripo's powerful 3D-native foundation models and Tripo API.

The resulting projects demonstrated just how broadly creators are beginning to interpret the idea of worldbuilding. Submissions included conventional games and virtual environments, but also children’s experiences, interactive fashion, physical devices, medical concepts, and applications connecting generated 3D content with real-time conversation.

The San Francisco event took place during SF Tech Week and brought together nearly 60 participating teams and more than 800 attendees. Most teams consisted of only one to three people, with participants coming from a mixture of creative, engineering, academic, and entrepreneurial backgrounds.

“We want to put the power to create into everyone’s hands,” said Tripo Co-Founder and Chief Scientist Dr. Yanpei Cao. “Many of the participants here today don’t have professional design or engineering backgrounds, and most are solo developers. But their projects have stood out alongside those of bigger, more experienced teams.”

The day also included discussions featuring researchers and representatives from Tripo, Runway, World Labs, Pika, Kling AI, and other companies working across interactive media, video, 3D generation, and spatial computing.

One panel considered how emerging tools fit into the broader creative process. The participants argued that creative software still needs a better understanding of the context in which directors, artists, and production teams collaborate. Their broader conclusion was that generating content is not equivalent to determining what should be made: human creators continue to provide the storytelling judgment, artistic direction, and understanding of an audience.

That distinction became particularly relevant during presentations about world models: systems designed to understand or generate environments that can respond to users rather than remaining fixed images or videos.

World Labs researcher Yi Hua presented Atlas, a multimodal world model developed to operate across text, images, video, and 3D data within a shared spatial context. Runway Research Director Jimei Yang discussed Solaris, an early research project that generates an interactive interface frame by frame. Instead of interacting with traditionally authored code, users provide clicks, drags, and text input that condition what appears next.

Dr. Cao of Tripo introduced Project Eden, Tripo's world model, and its latest milestone, EdenLive, a real-time generative renderer intended for rapidly prototyping interactive experiences. He argued that although agents are becoming more capable of producing code and assets, someone must still recognize an idea, explain it, evaluate the result, and decide whether it is worth developing.

“Everything started as an idea, but it may have been too much time or bother before,” Dr. Cao said. “Now, if we can describe it, we have tools that can turn a passing idea into a real thing.”

The projects presented during Tripothon offered more concrete examples of what these new workflows might look like.

Jennifer Jing received the San Francisco event’s Best Use of World Labs award for Agent Blindbox, a virtual collectible experience built around characters intentionally styled as “AI slop.” Players open a trash can to reveal different characters, each of which inhabits and interacts with its own environment.

The project turns the frequently dismissive phrase into part of its premise. Jing wanted to explore whether something labeled as disposable could become desirable when it was thoughtfully designed, packaged, and supported by a sense of mystery.

She used Tripo for image generation and editing, multipart 3D asset creation, character rigging, and animations including walking, running, and jumping. The environments were created through World Labs and embedded into the website, while Figma and coding tools were used to refine and assemble the final experience.

Several of the event’s most successful entries focused on VR and WebXR. UC Berkeley student Yichi Zhang placed third with Space Archaeology, a VR exploration game that sends players across the solar system to collect artifacts.

Second place went to Dulce Baerga’s Dear Little Me, a WebXR application that transforms children’s drawings into interactive 3D playthings. Tripo was used for 3D generation, rigging, and animation, while World Labs helped create the virtual playroom.

Baerga described the project as a whimsical imagination space for children and a potential shared experience for parents and kids. Rather than treating a drawing as a finished image, Dear Little Me uses it as the starting point for something a child can encounter and play with spatially.

The top prize in San Francisco went to former Ubisoft marketing manager Annette Lapham for Halls of History: The Veiled Study. The conversational mystery game allows players to speak with historical figures, search for artifacts, and defend the conclusions they reach.

Its rooms were built as Gaussian splats using World Labs’ Marble, while Tripo-generated objects populate the environments. More unusually, the Tripo API became part of the game mechanic itself: the system generates unique artifacts for players to hunt based on their conversations. Those objects can then be combined to solve a puzzle and unlock the next room.

Not every project remained inside a virtual environment. Gemtronic by Aspirium combined 3D creation with electronics and wearable design to produce a smart ring containing a digital pet.

The pet hatches from an egg, and changes form as the wearer walks. Buttons and hand gestures control the ring, which also includes games, clock faces, and a rhythm mode responding to surrounding music. Its electronic components remain consistent, but the external shell can be replaced to change the ring’s appearance.

Creator Linzhi Wang used Tripo to explore different shell shapes before taking selected designs into CAD software for further refinement and physical construction. In this case, generated models were not treated as finished manufacturing assets. They served as fast design candidates, allowing the team to compare multiple ideas before investing time in precise engineering.

Across the entire global competition, the first-place award went to Somehow Still Alive, a third-person survival shooter. Second place was awarded to the immersive experience A Long Farewell, while third place went to an application created by a student from Keio University that transforms children’s crafts into 3D creatures.

The projects vary significantly in technical ambition, visual quality, and intended audience, but that variety is arguably the most meaningful result of the event. “Worldbuilding” is no longer limited to manually constructing a static environment that every player experiences in exactly the same way. Creators are experimenting with spaces that respond to conversation, generate objects during play, connect children’s drawings to interactive experiences, or move from digital prototypes into physical products.

Whether world models become a practical component of mainstream production will depend on more than their ability to generate an environment. Artists and developers will still need control, consistency, reliable export options, optimization, and the ability to revise the results without rebuilding everything from scratch.

Tripothon nevertheless provides a look at how small teams are already testing these ideas. The most interesting projects were not necessarily those that generated the largest quantity of content. They were the ones that connected emerging 3D tools to a clear creative premise, interaction, or audience.

As the underlying technology becomes more accessible, that creative judgment may become even more important. Building a world is one challenge. Giving someone a reason to enter it remains another.

Tripo Team, Software Developer