Editorial illustration for Decart’s world model simulates hours of photorealistic driving
Decart’s world model simulates hours of photorealistic...
Decart’s world model simulates hours of photorealistic driving
Decart rolled out Oasis 3 on Wednesday, a real‑time world model that can render hours of photorealistic driving scenes. The startup is offering the system through an API, aiming first at autonomous‑vehicle firms that need to rehearse rare road conditions at scale. While the tech is impressive, Decart isn’t stopping at cars; it says robotics and other physical‑AI tasks are next on the roadmap.
Here’s the thing: by opening access from day one, Decart hopes to spark a developer community similar to what OpenAI did with language models. “It’s going to be the first usable world model that people can actually program on top of,” co‑founder and CEO Dean Leitersdorf told TechCrunch. The company already counts more than 100 000 developers in its ecosystem, many of whom are building tools on Lucy, Decart’s real‑time video model used in e‑commerce and live streaming.
Access costs $0.02 per second, with enterprise rates varying by use case. Decart joins a crowded field—Google’s Genie 3, World Labs’ Marble, and video‑generation firms Luma and Runway are all pushing physics‑aware world models forward.
Oasis 3’s release comes a few weeks after two-year-old Decart raised $300 million, which Leitersdorf says followed “huge demand increases for the models we built” in e-commerce, live streaming and physical AI.
Why this matters
We see Decart’s Oasis 3 delivering photorealistic driving scenes in real time, and the API‑first approach could lower barriers for teams building autonomous‑vehicle simulations. Yet the model arrives alongside Google’s Genie 3 preview, World Labs’ Marble service and video‑generation efforts from Luma and Runway, so the market is already crowded. For developers, immediate access means we can prototype rare scenario testing without building a full engine, but the brief note about “some caveats” leaves open how consistent the visual fidelity remains across weather.
Autonomous‑vehicle firms may appreciate scale, yet it is unclear whether the API pricing or latency will meet production‑grade demands. Robotics researchers might repurpose the physics‑aware rendering, but integration work could offset the promised speed gains. In short, Oasis 3 adds another tool to our toolbox; we should experiment carefully, track performance metrics, and stay aware that broader adoption will depend on concrete reliability data rather than early hype.
We’ll watch how it integrates.
Further Reading
- The Waymo World Model: A New Frontier For Autonomous Driving Simulation - Waymo Blog
- Decart AI Lab | Real-Time World Models - Decart AI Lab
- Researching Foundational Real-Time World Models - Decart AI Lab - Decart AI Lab
Related Reading
Industry Applications Verizon Acquires TracFone as More Brands Shift to MVNO Model
Industry Applications Company unveils 5-point Community-First AI plan to curb data center energy use
Industry Applications 15 AI & ML Presentations 2025 Highlight Law Uses and Limits of AI
From Archives AI‑enhanced lessons in Sierra Leone: teachers lead impact study
From Archives