Reactor, a San Francisco-based developer platform for real time generative video and world models, announced today that NVIDIA and Sapphire Ventures have joined its Series A round led by Lightspeed Venture Partners. The company did not disclose the financial terms of the round, but the expanded investor group signals rising demand for interactive AI infrastructure. Existing backers include WndrCo, Amplify Partners, Sky9 Capital, FPV Ventures, and additional investors.
Real Time Infrastructure Demand
World models are moving beyond video generation into systems that can simulate, understand, and interact with dynamic environments. As these models become larger and more computationally intensive, running them interactively requires significant GPU resources and infrastructure purpose-built for low latency inference. Reactor provides that infrastructure through a proprietary inference engine, a global GPU cloud, and a low latency streaming layer.
Unified SDK and API
Developers and enterprises can run models at more than 60 frames per second with latency under 40 milliseconds through a unified SDK and API. The platform abstracts the complexity of deploying and scaling real time systems, enabling teams to focus on application development. This capability is already attracting active projects from major Hollywood studios, advertising platforms, and video streaming services.
Robotics and Physical AI
Physical AI and robotics represent another natural application for the same infrastructure. Many robot policies and world models require more compute than can practically run directly on a robot or autonomous system. Reactor can run those models in the cloud while streaming observations and actions between the model and the physical system in real time.
Closed-Loop Simulation
Reactor can also run world models as closed-loop simulated environments, allowing robot policies to be trained and evaluated inside realistic, responsive generated worlds. Teams can use Reactor to serve, evaluate, and train robot policies and world models without building low latency inference infrastructure themselves. The platform supports both open source models and proprietary customer-owned systems.
Investor Perspective
Sapphire Ventures brings more than 15 years of experience partnering with enterprise companies as Reactor scales production deployments across industries. Anders Ranum, Partner at Sapphire Ventures, said enterprise teams building interactive video, gaming, or robotics applications eventually face infrastructure constraints, and Reactor's purpose-built platform solves that challenge. Early production deployments across media and robotics demonstrate the breadth of the opportunity.
Leadership and Market Momentum
Reactor was co-founded by CEO Alberto Taiuti and CTO Bryce Schmidtchen, both former technical leads on Apple Vision Pro. Taiuti previously co-founded Luma AI and served as CTO, helping build infrastructure for one of the world's most widely used 3D and video generation platforms. The broader team includes engineers and researchers from Apple, Meta, Google, Adobe, Replicate, and Microsoft.
Market Shift and Availability
Ahmed Ahres, Head of GTM at Reactor, said developers have moved from asking whether real time world models are possible to asking how quickly they can put them into production. Studios are using these systems in live production workflows, and robotics teams are exploring how to run sophisticated models without placing all that compute on the robot. Reactor is available today via SDK and API with usage based pricing that varies by model, and developers can get started at reactor.inc/signup.
Reactor's expanded Series A investor group reflects a broader market shift toward interactive AI systems that require specialized real time infrastructure. By combining a proprietary inference engine, global GPU cloud, and low latency streaming layer, the company positions itself as an infrastructure provider for media, robotics, and emerging physical AI applications. As world model adoption grows, the platform aims to let developers focus on product innovation rather than the cost and complexity of low latency deployment.