Fun to try as an interactive video demo, but developers face no public pricing and newer Odyssey models already exist.
Kenning's take
What it does
Odyssey-2 is a world model that generates video frame by frame and streams it straight away, about 20 frames a second, so a viewer can steer what happens by typing text prompts while it plays. Odyssey released it in October 2025 as a web experience. Its successor Odyssey-2 Pro is offered to developers through the Odyssey API, and the company now promotes Odyssey-3 as its newest model.
- Video streams start almost instantly
- A new frame every 50 milliseconds
- Steer the video with text prompts while it plays
- Multi-minute continuous video
- Odyssey-2 Pro available through API and SDKs
Who it suits. Developers and curious users exploring interactive AI video and world models.
Key facts
- Category
- AI models
- Made by
- Odyssey
- Released
- 27 October 2025
- How to get it
- try it in Odyssey's web experience; developers get Odyssey-2 Pro through the Odyssey API with a developer account
- Works on
- Web, API
FAQ
When was Odyssey-2 released?
Odyssey announced Odyssey-2 on 27 October 2025.
How do developers use Odyssey-2?
Odyssey-2 Pro is available through the Odyssey API, with JavaScript and Python SDKs, after creating a developer account.
Is there a newer Odyssey model?
Yes. Odyssey's site now promotes Odyssey-3 as its most capable world model.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
ACE-Step 1.5
Open-source music generation model that writes full songs with lyrics and runs on ordinary consumer GPUs.
AlphaGenome Atlas
Google DeepMind database of AI-predicted effects for every possible single-letter change in human DNA.
Amazon Nova 2
Amazon's second family of Nova AI models, covering reasoning, speech conversation and combined text-and-image generation.
Cisco Antares
Cisco's small open-weight models that find where known security vulnerabilities sit in a codebase, running locally.