China’s ByteDance has released Vidi2, an AI video editor that can take in hours-long footage, and use it to generate a TikTok video, or a movie. Deedy Das, Partner at Menlo Ventures claimed that the video editor “understands video better than even Gemini 3 Pro.”
Vidi2 is a multimodal large language model with 12 billion parameters designed specifically for video understanding. It can reportedly process hours-long raw footage, understand the narrative structure, and generate complete TikTok short videos or movie clips based on simple prompts. This has been seen as a major disruption to the existing video editing industry.
This new model relies on a fine-grained spatiotemporal localization (STG) feature that can simultaneously identify time stamps and bounding boxes of objects in the video. Given a text query, Vidi2 not only finds the corresponding time period but also accurately marks the location of specific objects within those time frames.
READ: US and China reach ‘final deal’ on TikTok sale (
ByteDance has developed multiple practical automated editing tools, including highlight extraction, story-aware cutting, content-aware layout reconstruction, and multi-angle switching, all of which can run on consumer-grade hardware, all based on Vid2’s capabilities.
The technology has been applied to TikTok’s Smart Split feature, which automatically edits, reconstructs, adds subtitles, and converts long videos into short clips suitable for TikTok. Vidi2 can also transform simple prompts or trending topics into structured video titles, openings, and outlines.
According to AIbase, the release of Vidi2 and ByteDance’s huge TikTok (1 billion daily active users) data platform advantage has given it massive video data for training and real-time feedback optimization, posing a significant challenge to native AI companies. As the technical flywheel of big platforms starts to turn, traditional AI companies may face greater competitive pressure.
READ: ByteDance eyes $330 billion valuation amid uncertainty over TikTok in US (
Vidi2 is still in the research phase, and a Demo will be released soon.
China-based ByteDance is best known for being the parent company of the massively popular social media platform TikTok. Recently there have been talks for the divestiture of TikTok’s U.S. operations from the parent company. U.S. president Donald Trump had signed an executive order last month to advance a deal on the TikTok app with a group of mostly American investors. This deal would allow the app to stay online, in accordance with a bipartisan law passed in 2024 which required its U.S. operations to be divested from the Chinese parent company.
ByteDance also recently launched an AI assistant for phones. This assistant will debut on ZTE’s Nubia M153 smartphone. The tool uses the Doubao large language model to handle spoken tasks such as content searches and ticket bookings. ByteDance confirmed discussions with multiple manufacturers to integrate the assistant into future smartphones. The firm stated it has no intention of developing its own hardware.