Introducing Oasis: the first playable AI-generated game. We partnered with @DecartAI to build a real-time, interactive world model that runs >10x faster on Sohu. We're open-sourcing the model architecture, weights, and research. Here's how it works (and a demo you can play!):

Join the hackathon and apply here:

We believe today's models significantly underutilize inference-time compute and search. With better algorithms and much faster hardware, we can push this scaling law by several orders of magnitude. We're particularly interested in research that accelerates parallel reasoning,.

Each team accepted gets a free 8xH100, exclusive access to Cognition's API, and Anthropic credits for 24 hours. We'll pick the winner by the best application of inference-time compute, judged by a panel of researchers from major AI projects.

We're excited to partner with @Cognition_Labs @Mercor_AI @CoreWeave and @AnthropicAI to host an inference-time compute hackathon, featuring >$60K in cash prizes and >1 exaflop of free compute.

RT @DecartAI: Thousands of people in the line 😅 adding compute to handle the load!.Go to to try it yourself https:/….

Oasis was trained on purely open-source data from VPT, OpenAI's open-source Minecraft dataset (MIT license)! You can learn more about that dataset here:

Check out the open-source version (model weights and code) here: You can learn more about that dataset here:

Tens of thousands of people are using Oasis, the world's first AI-generated game. Try it yourself at . It's trained on purely open-source data from VPT, OpenAI's open-source Minecraft dataset (MIT license)!.

@DecartAI @julianhquevedo @qamcintyre @sprooos @endernewton You can play Oasis here: Learn more about the model and partnership here:

You can play Oasis here: Learn more about the model and partnership here:

Thanks again to the @DecartAI team for the partnership and for @julianhquevedo @qamcintyre @sprooos @endernewton for bringing Oasis to reality!.

Gameplay is just the start. Soon, most of the internet will be AI-generated, and >70% of internet traffic is video. Video diffusion models are extremely compute-bound, needing to processes thousands of tokens per frame. This workload is perfect for Sohu.

We built Oasis to run incredibly fast on our upcoming AI chip, Sohu, featuring a DiT backbone, a ViT auto-encoder, and more (see our technical report). Sohu is built to handle massive video models, running in 4K for 100B+ parameter models.

Oasis generates frames based on your keyboard inputs. You can move and jump around, break blocks, and build and explore a brand new map every game.

Today we announced our $120M fundraise to bring Sohu to the world. Thanks @edludlow for hosting us!

RT @johncoogan: INTERVIEW WITH FOUNDERS OF @ETCHED.the thiel fellows fighting nvidia. - @UbertiGavin & @robertwachen on AI chips.- Debating….

Thank you to all our supporters - @peterthiel, @ashtom, @jasoncwarner, @amasad, @kvogt, @balajis, @kevinhartz, @immad, @bryan_johnson, @amiruci, @novogratz, David Siegel, Stanley Druckenmiller and many more.

Transformers have a huge moat. We believe in the hardware lottery: the architecture that wins is the one that runs fastest and cheapest on hardware. Transformers won the lottery: AI labs have spent hundreds of millions of dollars optimizing kernels for transformers. Startups use.