Report
Om AI bets on understanding real video instead of generating it
Tech in Asia reports that the Hangzhou-based startup is building lightweight vision models for creators and robots.
TLDR
Tech in Asia reports that Om AI is focusing on video understanding rather than generative video. It says the startup’s small models can analyze hundreds of gigabytes of footage locally, captioning and tagging it and making video libraries searchable without cloud processing.
Combined views
1K
1 Source, first seen 7h ago
