Framy v2.0 — Up to 30 Seconds of Video in One Take

Framy v2.0 is out — a new version of the video model. You can now generate a clip from text, from a first frame, or between a first and a last frame.
Length is 3 to 30 seconds per generation. No more stitching a long scene together from short pieces.
Resolution is 720p or 1080p. Aspect ratio is 16:9, 9:16, or 1:1.
The model holds fast motion, tracking shots, and camera orbits more steadily: characters no longer "melt" mid-shot.
Sound is generated together with the video, in stereo. Lips match the lines more accurately.