Odyssey-3
Odyssey-3 is a foundation world model that creates generated environments from a prompt and predicts their changes in real time as users or agents interact, with advanced physical accuracy. It supports applications like controlling robots, humanoids, and vehicles, and is offered via research preview and API access to developers building physical AI systems.
Ratings & Reviews
Write a reviewBe the first to share a thoughtful review.
Information
Website checks
Automated checks of the public website by Google Lighthouse 13.5.0. Scores can vary between runs.
Checked Oct 22, 2026
Test this website on PageSpeedWebsite checks refresh automatically each week.
- Performance
- 54 out of 100, Needs improvement
- Accessibility
- 94 out of 100, Good
- Best Practices
- 96 out of 100, Good
- SEO
- 92 out of 100, Good
Around the Web
It’s a cool shift, getting AI to predict how the world reacts rather than just describe what it sees.
henry_habib · Imported from Product Hunt
Best-of-8 is doing a lot of work in that 66.1. If the scores it's being compared against are single sample then it isn't the same measurement, and real-time use is single sample by definition because you only get one rollout when you're acting in the environment. The distilled few-step model's single-shot score is the number I'd want on the page. Publishing that would be a stronger claim than the best-of-8 headline, not a weaker one.
asadmalik901 · Imported from Product Hunt
the jump from "generates video that looks physically plausible" to "controls a real robot arm" is the part I'd want more detail on. a world model can score well on Physics-IQ purely on visual plausibility while still being wrong about the specific numbers that matter for control, friction coefficients, exact mass, where contact actually happens. when you adapted this to robot arms and a car, was the visual model retrained on real sensor data too, or is the same video-prediction core driving control decisions directly
galdayan · Imported from Product Hunt