Tongyi Video Generation Model Wan2.7-Video Launch: Supports full-modal input of text, images, videos, and audio

Phoenix.com Tech News: On April 3, the video generation model Wan2.7-Video has officially launched. The model supports full-modal input of text, images, video, and audio. Focusing on editing and modification capabilities after video generation, it aims to make video content as editable as documents.

In video editing, Wan2.7 supports making localized adjustments to the visuals through instructions, without needing to regenerate the entire clip. Specific functions include: adding or removing elements (such as “delete the train in the video”); replacing objects (such as “swap the film reel for a plate”); modifying object attributes (such as changing the building color); and supporting precise additions based on reference image content. On the environment and style side, while keeping the characters’ actions unchanged, it can change the background season from summer to deep autumn, or convert to artistic styles such as feltcraft (wool felt). In addition, it also supports improving video quality (such as black-and-white colorization), visual understanding tasks (such as subject segmentation), and adjusting the shooting method (such as modifying the focus).

In terms of plot modification, Wan2.7 allows modifying character dialogue, actions, and even the shooting perspective without changing the original identity and setting. Dialogue edits can keep emotion, mouth movements, and timbre consistent; action edits such as “the girl sitting on the couch turns into someone standing and playing games.” At the story level, the player character can be replaced with a medieval knight while maintaining the original gripping posture. Parameters such as camera position, shot type, lens type, focal length, and more can all be adjusted via instructions—such as “modify the camera to gradually rise upward from the ground.” These functions are designed to lower the technical barrier for video secondary creation and provide creators with more flexible control tools.

(Editor: Dong Pingping)

     【Disclaimer】This article only represents the personal views of the author and is not related to Hexun. The Hexun website maintains a neutral stance toward the statements, viewpoints, and judgments contained in the text, and does not provide any express or implied guarantees regarding the accuracy, reliability, or completeness of the content included. Please readers use this information only as a reference and bear full responsibility for your own actions. Email: news_center@staff.hexun.com

Report