- Runway is among the first to demonstrate video generation models on NVIDIA Rubin platform. Runway’s Gen-4.5, the world's top-rated video generation model, was ported from NVIDIA Hopper to Vera Rubin NVL72 in a single day, demonstrating seamless backward compatibility and production readiness.
- Runway's first general world model family, GWM-1, also exemplifies the applications Rubin is designed to accelerate: from physics-aware robotics training to explorable virtual worlds and interactive avatars.
We are accelerating our video and world models with the NVIDIA Rubin platform. Gen-4.5 is the first video model to run on this next generation of AI infrastructure.
As video generation models produce exponentially longer sequences, demand on the compute infrastructure grows dramatically. Meeting this demand requires a purpose-built AI infrastructure, designed for the scale, efficiency, and memory capacity that video diffusion requires.
The Rubin platform is built for the next frontier of AI models, including video generation. With 50 PF of inference compute per GPU, the platform accelerates real-time, long-form, high-fidelity video generation. Unlocking new creative and commercial possibilities for users worldwide.
Runway’s Gen-4.5, currently the top-rated video generation model in the world, was ported from NVIDIA Hopper to Vera Rubin NVL72 within a single day. This rapid integration demonstrates strong backward compatibility across NVIDIA GPU generations and confirms Rubin's readiness for production video generation workloads at scale, powering the next frontier of AI: systems that move beyond text to understand, simulate, and interact with the physical world.
The recently launched GWM-1, Runway's first general world model family, exemplifies the workloads Vera Rubin was built for. GWM-1 simulates physics-aware environments in real time for robotics training, explorable virtual worlds, and interactive avatars.
These applications require sustained, long-context inference that Rubin is uniquely positioned to accelerate. By enabling AI systems that can reason and generate coherent representations of the physical world from text or image inputs, Rubin opens the door to breakthrough applications across entertainment, physical AI, and beyond.
“GWM-1 is our first step toward models that don’t just generate pixels. They understand and simulate the world behind them.” said Cristóbal Valenzuela, Runway co-founder and CEO, “These are long-context, physics-aware workloads, and that’s exactly where NVIDIA Rubin platform shines. Together, we’re accelerating a new class of world models that can power explorable worlds, interactive avatars, and robotics training. Opening the door to breakthrough applications across entertainment, physical AI, and beyond.”
"Video generation and world models are ushering in a new era of AI, one that understands and simulates the physical world," said Richard Kerris, Vice President and General Manager of Media at NVIDIA. "Vera Rubin was architected from the ground up for these demanding workloads, and Runway's rapid integration of Gen-4.5 shows our platform is ready to power the next generation of creative and physical AI."
Runway, powered by NVIDIA, is building the infrastructure for AI systems that move beyond text to understand, simulate, and interact with the physical world.
