Today Higgsfield, a multimodal AI company, announced the release of Diffuse 2.0 — host of new and updated AI video generation features that "removes the burden of production and expedites personal storytelling to minutes instead of days.”
Diffuse sells itself as an entertainment app that specializes in using text-to-video AI to generate meaningful stories and videos. Diffuse 2.0 builds on this framework by combining the self-discovery features of TikTok with the storytelling content found in choices/episodes to help create more dynamic videos. Alongside this new design is several new and upgraded features.
The Selfie Animator is a full-featured photo animation tool that preserves the subject's face, clothing, and background. The new video templates add life and motion, while the Stories feature places users into scripted stories with AI visuals.
Feeling snacky?
Diffuse 2.0 also includes a feature called Video Snacks, which uses details provided by the user to create personal vignettes and video-to-video stylization which can turn your video into several different styles. In other words, Diffuse 2.0 finally has a Text-to-video builder — making creating AI videos from scratch easier than ever.
This tool might not be on everyone’s radar just yet, but, since its launch in 2024, it has amassed one million followers, 50% of which are sharing their content with others.
“Video represents 80% of web traffic, and has become the de facto way for people to express themselves," Higsfield AI creator Alex Masharbov commented. "But it’s difficult, expensive, and time consuming to create a high quality video – until now."
Currently Diffuse is available on both iOS and Android, however it will require a subscription for continued use. For more information bout the changes you can follow Higsfield AI on X.
If you are more of a fan of image generation we have a great list of the best AI Image Generators currently available and a comprehensive list of the best AI tools and devices of 2024.
More from Tom's Guide
