Hsuan Su

@chrmanning and I went on @latentspacepod to talk about world models. youtu.be/oBWRHnggscM?si…

One convergence that is increasingly clear: games provide a natural incentive for humans to interact with complex environments, generating trajectories that are valuable for learning intelligent behavior.

We refreshed our logo to better reflect what we are building: multimodal world models that simulate environments, predict outcomes, and enable planning over long horizons. We kept the moon from our original mark and evolved it into a symbol of depth, generation, and evolving worlds.

Creators, worldbuilders, game devs, Today we’re launching the first ever Moonlake Creator Cup. It can be a game, a simulation, a story or a tiny world of your own. Anything you imagine. We’ll award $30,000 across the most creative worlds, including $10,000 for first place. If you’ve been curious about Moonlake, this is a good excuse to build something. Repost “Moonlake Creator Cup” to join. Let’s see some magic ✨

Moonlake is here at GDC 2026 in San Francisco 🌉
Doors open in 10 minutes. Good food, drinks, and a beautiful venue tonight. DM me if you want to come by :)

:)

this is a lot more practical than Genie!

We releasing an early version of Moonlake's world modeling agent. It is not just a tool for generating beautiful scenes. It is a system for creating worlds you can interact with. Objects respond. Physics runs determinstically. Your actions have real causal effects. We believe in building with the community, not in isolation. Over the past month, we have iterated closely with users, hosted hackathons, and worked side by side with hundreds of beginner game developers to understand their workflows, frustrations, and creative ambitions firsthand. Our intention is simple. Help developers solve real pain points. Expensive tooling. Repetitive tasks. Tedious work that slow down creativity. We believe humans are the true source of creativity. Tools should amplify that power, not replace it. Game developers, and more broadly creators, will always be at the center of our ecosystem. This is our first beta release. It is an early step, but we hope it gives you a clear sense of what we are building toward. A future where anyone can create rich, interactive worlds with intention and meaning.

All generated directly through our prompt-based chatbox: – 3D assets – Characters – Animations – Game mechanics – Scene generation – Skyboxes Our web app is now rolling out, come try it for yourself.

A closer look at Cyber Strike, a game built in our desktop product by a previous hackathon winner using prompt-driven creation and granular engine edits. Time Stamps: (0:00-0:35) - Introduction (0:36-3:35) - Gameplay (3:36-4:28) - Ending notes

Game generation → diffusion output, end-to-end. From prompt to scene: a moon-like environment created in our engine and reskinned into flexible media assets. Built on our web app.

“Riftwater”, a sci-fi 4-dimensional fishing game built on our web app in 2 days. Prompted entirely through natural language: – Mechanics – Assets – Physics – NPC behavior – Encounter Rolls/ RNG - Sounds and EFX

Join the founders of @moonlake to learn about and engage in building real-time interactive worlds using their spatial reasoning models that can vibe code worlds! Jan 14, 6pm, SF. partiful.com/e/JbKr33S4uNPY…

Join us next Wednesday, Jan 14, in San Francisco for a game jam hosted with @a16z @AICollectiveCo. Explore a new way to prototype and build games. RSVP → partiful.com/e/JbKr33S4uNPY…

More looks at Reverie v0.1.0 Prompts → Mechanics → Interaction Author game mechanics through prompt-driven interactions and direct your game’s vision without being slowed by repetitive backend work. In frame: one scene, six different interactions.

Faculty Recruitment at National Taiwan University AI Center of Research Excellence (NTU AI‑CoRE) Start date: August 1, 2026 Apply online (rolling review; early applications encouraged): academicjobsonline.org/ajo/jobs/31462 Learn more about the Center: ntu.edu.tw/spotlight/2025…

Reverie v0.1.0 vs MirageLSDv2 How we're building world simulators: moonlakeai.com/blog/why-world…

1/ Today, we are thrilled to announce Generative Simulators, a new class of adaptive, auto-scaling environments for AGI training and evaluation 🤖🧵 Static datasets, hand-authored environments, and human-curated demonstrations do not automatically scale with the learning patterns of the trained model. We propose Generative Simulators as a principled alternative: environments that evolve, evaluate, and adapt to agent behavior over time. Technical Report: patronus.ai/generative-sim… Blog: patronus.ai/blog/introduci…