Luma AI, the frontier artificial intelligence company building multimodal AGI, has announced the latest update to its Ray3 reasoning video model Ray3.14. The new version is intended for animation, video-to-video, and cinematic workflows, with an emphasis on temporal stability, motion fidelity, and consistency.
Ray3.14 has been developed for professional creative environments, offering native 1080p outputs, generation speeds up to four times faster, and per‑second pricing at one‑third the previous cost. It is designed to support animation and high‑fidelity workflows, addressing issues such as flicker, drift, and inconsistency that affect other models.
Ray3 introduced reasoning‑based video generation by interpreting scenes as a whole and maintaining coherence across motion, lighting, characters, and camera behaviour. Ray3.14 applies this reasoning engine more effectively to animation and professional video workflows, producing greater detail adherence and enabling creative teams to move from experimentation to execution with confidence.
“Ray3.14 is designed for creators who need animation and video to behave like real production assets. By delivering native 1080p, dramatically faster generation, and per-second pricing that is three times cheaper, we’re giving advertisers and filmmakers a model that excels in animation and can be trusted for real-world creative workflows,” said Luma AI CEO and co-founder Amit Jain.
Ray3.14 is designed for environments where speed, reliability and creative control are essential for deployment at scale. The update introduces:
- Quality and stability for animation and video‑to‑video workflows: Delivers Luma’s highest level of visual stability to date, maintaining temporal coherence and motion fidelity. Characters, environments and styles remain consistent across frames, supporting narrative continuity and production‑ready outputs.
- Native 1080p across core workflows: Generates video natively at 1080p, removing the need for post‑upscaling. Outputs are suitable for broadcast, streaming and digital delivery, moving directly into editorial and distribution pipelines.
- Speed for creative iteration: Provides generation speeds four times faster, with reduced time‑to‑first‑frame and end‑to‑end output. Enables agencies and filmmakers to explore more concepts, compare options side by side and cut down on revisions.
- Per‑second pricing for production scale: Introduces costs three times lower for typical outputs, making campaign‑scale use cases such as multiple formats and regional variations more viable. Offers predictable costs and supports scaling without increased budgets.
