PixVerse launches C1 video model for film
PixVerse has released C1, its first video model built for film production, aimed at the consistency, physics and multi-shot problems that have kept AI video out of professional production pipelines.
PixVerse announced C1 on 7 April 2026 in Singapore. The company calls it its first AI video model purpose-built for film production. It puts three things into one model: an action engine, a visual effects system and multi-panel storyboarding.
According to PixVerse, C1 targets the kinds of sequences where AI video has usually broken down, including close-combat choreography, high-velocity motion and object interactions. The company says these are rendered with consistent spatial relationships and physical weight, which reduces manual frame-by-frame correction. It also says the model supports particle and lighting effects, fluid dynamics and atmospheric environments, in both realistic and stylised scenes.
A new storyboard-to-video feature turns uploaded static storyboard layouts into continuous video sequences. Scenes are split automatically based on how the prompt is structured. This works across text-to-video, image-to-video and reference-guided modes. Reference-guided generation is meant to keep characters and scenes consistent from shot to shot. PixVerse Canvas, a visual workspace, sits alongside the model for planning sequences before generation.
C1 generates video at up to 1080p and up to 15 seconds long, with native audio. It is available now through PixVerse Web and through an API for enterprise integration.
The announcement includes no benchmarks, independent evaluations or pricing. PixVerse says its platform has more than 100 million users across 175 countries. Its other models include the flagship V6 and R1, a real-time world model launched in January 2026.