Skip to content

VideoOdyssey

Odyssey

Real-time world model that generates interactive, explorable AI video.

Categories
VideoGaming
Pricing
FREE
Hosting
Cloud
Platforms
Web
Models
Self-contained (on-device)
Verified
Jun 15, 2026

Odyssey is a world-model startup whose AI generates interactive video you can steer in real time — explorable, 3D-like scenes streamed frame by frame without a game engine. Its current model, Odyssey-2, turns a text prompt or image into a multi-minute simulation you navigate as it is generated. It is positioned for film, gaming, and interactive media, and is available as a free research-preview experience.

Pros & cons

  • Real-time interactive generation
  • Steerable, multi-minute scenes
  • No game engine or assets needed
  • Free research-preview access
  • Research preview, not production-ready
  • Limited resolution / visual fidelity
  • No commercial terms or pricing yet

Tags

Further reading

View all Video
  • View Decart details
    GamingPAID

    Decart

    Decart

    Real-time generative world and video models (Oasis, Lucy).

    Decart is an AI lab building real-time generative world and video models, including Oasis (a fully AI-generated, playable Minecraft-style open world) and Lucy (real-time video transformation used in gaming, ads, and virtual try-on). Its Oasis 3 is positioned as an API-accessible world model for Physical AI, with a low-latency inference stack. Products are accessed via the platform API and interactive demos.

    Worth knowing

    Raised $32M at a $500M valuation (Dec 2024, Benchmark), then $300M at ~$3.1B (Aug 2025); founders came out of IDF Unit 8200.

    • gaming
    • world-models
    • real-time-video
    • generative
  • View Marble details
    3DFREEMIUM

    Marble

    World Labs

    Generate editable 3D worlds from text, images, or video.

    A generative world model that turns text prompts, photos, videos, panoramas, or 3D layouts into editable, downloadable 3D environments. The first commercial product from World Labs, the spatial-intelligence lab co-founded by Fei-Fei Li. Outputs can be refined and exported for use in game engines and 3D pipelines.

    Worth knowing

    The first commercial product from World Labs, Fei-Fei Li's spatial-AI lab that hit unicorn status months after its 2024 founding.

    • world-model
    • 3d-generation
    • spatial
    • scene-generation
  • View Runway details
    VideoFREEMIUM

    Runway

    Runway

    Professional video AI. Gen-series models + a full editor.

    End-to-end video AI platform — text-to-video, image-to-video, in-painting, motion brush, and a timeline editor. Longest-running studio in the space; default choice when the output ships to clients.

    Worth knowing

    Co-created the original Stable Diffusion in 2022; raised a $315M Series E in 2026 at a $5.3B valuation.

    • video-gen
    • editing
    • motion
    • film
  • View Luma Dream Machine details
    VideoFREEMIUM

    Luma Dream Machine

    Luma AI

    Mainstream-quality video generation. Indie creator default.

    Luma Labs' video model surfaced as an iOS app and web tool. Strong text-to-video, image-to-video, and keyframe controls — the indie creator's go-to before reaching for Runway's heavier toolkit.

    Worth knowing

    Went viral at its June 2024 launch, hitting 1M users in four days with no marketing; Luma later raised a $900M Series C.

    • video-gen
    • image-to-video
    • keyframes
    • mobile