LinkedIn and 3rd parties use essential and non-essential cookies to provide, secure, analyze and improve our Services, and to show you relevant ads (including professional and job ads) on and off LinkedIn. Learn more in our Cookie Policy.
Select Accept to consent or Reject to decline non-essential cookies for this use. You can update your choices at any time in your settings.
Odyssey is an AI lab pioneering interactive video models—models that dream video you can interact with in real-time. This new form of general-purpose intelligence will power the next generation of gaming, education, companionship, simulation, and advertising, and unlock entirely new consumer, enterprise, and intelligence applications.
The dawn of pixel intelligence.
Odyssey-2 is a glimpse of video that feels sentient.
You type and intelligent pixels react instantly.
Soon video will talk to you, play music with you, teach you, help you learn new languages, improve your health, and much more.
Total sci-fi 🫶
Next-word prediction has taken us a long way. Where will next-frame prediction take us?
At Odyssey we believe it’ll become a general-purpose world simulator. Something radically new.
If you’re headed to NeurIPS and want to see a glimpse, email neurips@odyssey.ml.
Odyssey-2 generates video instantly. Not quick. Not fast. Instant!
Click "start streaming" and AI-imagined video appears on your screen in less than a second.
All for free.
Odyssey-2 now streams 16:9 video if you're on a big screen, and shows an awesome advantage of interactive video models.
Unlike pre-recorded video, real-time generated video intelligently adapts to the screen, viewer, and input device.
Really sci-fi when you think about it!
Odyssey-2 is live! It’s instant, interactive AI video.
You type a few words, hit “stream”, and video you can interact with is instantly imagined by AI.
It's so fun, and feels like magic.
We’re on a path to video of any kind being imagined by a model in real-time.
Because it’s real-time generated, each video becomes fully interactive & personalized. Type or speak into the model, and the video will respond in real-time.
Interactive video models are early, but the rate of advancement feels like early video models leaping from V1 straight to V3. Even in just the last 6 weeks we've seen insane leaps internally at Odyssey. Also Genie 3! We're going to see more major labs enter this domain soon.
This gets really wild if you consider any pixels on a screen to be video. Operating systems and apps could transition to being imagined by a model in the moment, fully interactive and personalized.
The infra required to stream video at this scale is substantial, likely exceeding Netflix. But, with zero video production cost (because the video is generated in the moment!), the unit economics are super compelling. Plus, video consumption likely goes way up.
Real-time video is a radical shift. It unlocks cinematic and educational experiences we only dream of today. Ask Da Vinci to sketch your idea as you describe it and see it take shape in real-time. Ask Einstein to explain a complex concept and interrupt him to ask a question.
We're bringing this new form of media to life at Odyssey, and I haven't felt this energized about anything, ever. We're hiring. Build magic with us! ✨