Open-source LongCat-Avatar turns one photo plus audio into minutes of lip-synced talking video

Roger_M_Taylor · x · 2026-09-15

A viral post highlights LongCat-Avatar, an open-source and free project that generates minutes-long, lip-synced talking-head video from just a single photo and an audio clip — capabilities that previously required a camera, studio and editing. The poster claims it "humiliated the entire paid AI video industry," with the repo link shared in the comments.

Original post →

More from Multimodal

Multimodal channel →