Published by Video Runtime · Source check 2026-09-14 · Evidence method · Suggest a correction
EDITORIALLY VERIFIED PRODUCT INFORMATION
About Odyssey-1
A public research preview of a narrow-domain playable world model that streams newly generated frames. Predicts the next visual state from current pixels, action input and previous states, then feeds generated output back into the autoregressive context.
Key features
Provides a public playable experience and openly describes its narrow training distribution and instability tradeoffs.
Getting started
Open the official project page and confirm current eligibility, region, account and payment requirements. Begin with the provider's own preview or onboarding flow; availability can change after this source check.
Real-time interaction
Predicts the next visual state from current pixels, action input and previous states, then feeds generated output back into the autoregressive context.
Pricing
Research preview; no standalone price verified Research preview; no standalone price verified Verify the provider's current account, region, quota and billing terms before starting a paid or production session.
Limitations
Odyssey itself describes the preview as raw, unstable and limited; long autoregressive rollouts can drift outside the training distribution.
EDITORIAL VIEW
Editor's Verdict
Odyssey-1 is relevant to real-time generative video, interactive worlds or continuous visual agents, with access limits recorded from its official source.
Treat preview footage and provider benchmarks as evaluation leads, not proof of availability or performance for your workload.
How It Works
- Predicts the next visual state from current pixels, action input and previous states, then feeds generated output back into the autoregressive context.
What We Like
- Provides a public playable experience and openly describes its narrow training distribution and instability tradeoffs.
Current Limitations
- Odyssey itself describes the preview as raw, unstable and limited; long autoregressive rollouts can drift outside the training distribution.
Pricing & Access
- Research preview; no standalone price verified
- Verify the provider's current account, region, quota and billing terms before starting a paid or production session.
Verification Summary
- Official provider material reviewed on 2026-09-14. Video Runtime did not create an account, pay, or run a performance test.
- All speed, quality, scale and price statements are attributed to the provider and are not independent measurements.
Technical notes & official performance
Official Performance Data
- Odyssey reports new frames every 40 ms and coherent streams for five minutes or more. These are provider claims from the 2025 preview.
Technical Notes
- The preview is post-trained on a smaller set of densely covered locations to trade generality for more stable rollouts.
Sources
- Official project or product page ↗official-site · accessed 2026-09-14
- Official documentation ↗official-docs · accessed 2026-09-14
Access, stage & mechanism
These are separate source-checked facts, not a live uptime monitor. Unverified fields are left open; transport and persistent sessions do not establish how frames are generated.
- Release stage
- Not verified
- Access mode
- Not verified
- Source availability
- Not verified
- Transport
- Not verified
- Interaction
- Not verified
- Session mode
- Not verified
- Generation mechanism
- Not verified
- Continuity method
- Not verified
- Licence status
- Not verified
- Tasks
- Not classified
CONTINUE EXPLORING
Related projects & models
Selected by research links and category, not by popularity.
A keyboard-controlled open-world research experience whose visuals and mechanics are generated by a world model.
Another frame streaming option in this section.Research demo, code, and model weights are linked
Create and navigate generated environments in real time through the experimental Project Genie experience.
Another frame streaming option in this section.Experimental Project Genie access; account, product and regional conditions may apply
A research preview for steering generated audiovisual worlds through camera movement and text actions.
Another frame streaming option in this section.Research preview; no general self-serve production access was verified
An experimental interface for steering objects and camera motion while AI video is being generated.
Another frame streaming option in this section.Public research preview; no production API or commercial release verifiedRead how official-source and hands-on records stay separate in our Methodology.