MUMBAI — Luma AI announced the release of Ray3.14, an updated version of its Ray3 video generation model, aimed at professional animation and video production workflows.
According to the company, Ray3.14 supports native 1080p video output, generates content up to four times faster than previous versions, and introduces per-second pricing that the company says is approximately three times lower for typical short-form video. The update is positioned for use cases where consistency across frames, motion accuracy, and visual stability are required, including animation and video-to-video applications.
Luma AI said Ray3.14 builds on the reasoning-based approach introduced with Ray3, which is designed to interpret scenes holistically and maintain coherence across elements such as lighting, camera movement, and character motion. The company stated that the updated model applies this approach more strongly to professional workflows, with an emphasis on reducing visual artifacts such as flicker or drift during longer or more complex sequences.
“Ray3.14 is designed for creators who need animation and video to behave like real production assets,” said Amit Jain, chief executive officer and co-founder of Luma AI. He added that the combination of higher resolution, faster generation, and revised pricing is intended to support real-world creative production needs.
The company said the model is optimized for production environments where speed and reliability affect deployment at scale. Ray3.14 generates video at 1080p without post-upscaling and is intended to integrate directly into editing and distribution pipelines for broadcast, streaming, and digital platforms.
Luma AI also said the updated pricing structure is designed to support campaign-scale video production, including multiple versions and regional variations, while maintaining predictable costs.
