Ray foundation video synthesis (Ray 3.2), UNI multimodal intelligence, and neural radiance 3D camera pathing.
Luma AI is a visual intelligence research company building generative foundation models, bridging neural 3D scene reconstruction and high-temporal-coherence video generation through Ray 3.2 and the UNI multimodal architecture.
Official model variants, context limits, and target filmmaking workflows
Current flagship video foundation model with enhanced physical realism, camera kinematics, and high temporal coherence.
Current multimodal foundation model supporting high-fidelity image synthesis and neural 3D scene representation.
Prior generation video foundation model developed by Luma AI.
Initial public generative video foundation model milestone developed by Luma AI.
Relationships mapped to canonical tools, techniques, prompts, workflows, films, and research
Every factual capability, benchmark, and parameter is backed by a verifiable primary source
Luma AI, Inc.