Pioneering generative video research, Gen-3 Alpha, Act-One character performance, and cinematic toolsets.
Runway is an applied AI research company advancing multi-modal generative video models for film, animation, and VFX. Known for Gen-3 Alpha, Act-One facial performance capture, Motion Brush, and the annual AI Film Festival (AIFF).
Official model variants, context limits, and target filmmaking workflows
High temporal fidelity, photorealistic video diffusion foundation model with advanced cinematic camera control.
Markerless expressive facial performance capture mapping camera-recorded actors onto stylized or photorealistic AI characters.
Embedded official demonstrations and filmmaker masterclasses
Official demonstration of Act-One architecture translating subtle micro-expressions, eye darts, and vocal articulation from actor footage onto non-human and cinematic stylized avatars.
Verified analysis and production case studies
A complete analysis of horizontal pan, zoom velocity curves, and vertical crane prompting syntax for cinematic narrative continuity.
Director's production dispatch on how Elena Rostova combined Midjourney styleframes, Runway camera motion, and Topaz 4K theatrical mastering to achieve a seamless dystopian cyber-noir aesthetic.
Relationships mapped to canonical tools, techniques, prompts, workflows, films, and research
Every factual capability, benchmark, and parameter is backed by a verifiable primary source