Meta Platforms unveiled a generative AI (genAI) tool capable of rendering high-quality videos based on text and image prompts, rivalling OpenAI’s Sora.

Named Movie Gen, the company said the system is powered by “the most advanced media foundation models” and is able to generate videos up to 16 seconds long based on written and photo inputs.

The product, which “outperforms similar models in the industry”, can also edit existing videos.

It has not been made available commercially to apps users, with Meta Platforms pushed the offering as being “part of our long and proven track record of sharing fundamental AI research with the community”.

The social media giant also referred to Movie Gen as its “third wave” of genAI work, combining images, videos, audio and 3D animation models, explaining it used a mix of “licensed and publicly available datasets” to train the system.

As for the tool’s editing capabilities, Movie Gen can remove, add and replace certain visual elements or make a more dramatic alteration such as background or style modifications, setting it apart from “traditional tools that require specialised skills or generative ones that lack precision”.

The company hopes “one day in the future, everyone will have the opportunity to bring their artistic visions to life and create high-definition videos and audio using Movie Gen”, noting it does not aim to replace artists and animators but “to help people express themselves in new ways”.

Meta Platforms hinted at integrating the offering with its social platforms Instagram and WhatsApp in the future.