Reactor Raises $59M to Build Infrastructure for Real-Time Generative Video

Reactor, a San Francisco-based developer platform for real-time generative video, raised $59M in combined Seed and Series A funding led by Lightspeed Venture Partners, with participation from WndrCo, Amplify Partners, Sky9 Capital, FPV Ventures, Abstract Ventures, and additional investors.

Founded by former Apple Vision Pro technical leads Alberto Taiuti and Bryce Schmidtchen, Reactor is building infrastructure for developers working with world models and real-time AI applications. The funding highlights a growing shift in venture capital toward the infrastructure layer supporting AI deployment, not just the models themselves.

Reactor just raised $59M, and the real story isn't the capital. It's the timing. While the market debates which models will win, Reactor is focused on the infrastructure required when developers move from experimentation to building products people actually use. The combined Seed and Series A was led by Lightspeed Venture Partners, with participation from WndrCo, Amplify Partners, Sky9 Capital, FPV Ventures, Abstract Ventures, and additional investors. More than a funding round, this is a bet on the infrastructure needed to support real-time AI applications.

What Happened

Reactor's origin story makes the funding easier to understand. Co-founders Alberto Taiuti and Bryce Schmidtchen, both former technical leads on Apple Vision Pro, spent roughly 1 year exploring what real-time video could become when world models moved beyond research environments and into production. Instead of building another model, they focused on the tooling developers need to put those models to work.

The company emerged from roughly 6 months in stealth and already has hundreds of early developers using the platform. That's a meaningful signal because developers care less about headlines and more about whether a product solves a problem, scales reliably, and saves time. Reactor provides SDK and API access for real-time generative video, allowing developers to stream generated pixels, audio, and actions with minimal implementation effort. The goal is simple: become the connective layer between frontier world models and the teams building commercial applications around them.

Why This Matters

The AI market has become obsessed with models. Bigger models, faster models, newer models. That's where attention naturally gravitates because models are visible, while infrastructure rarely gets the spotlight. Yet infrastructure is where many technology markets ultimately create durable value. Developers can admire breakthroughs all day long, but products reach customers only when somebody solves deployment, scalability, reliability, latency, and usability.

Reactor is positioning itself inside that gap. Rather than competing with frontier model labs, the company wants to help developers build on top of them, which is a fundamentally different strategy and one that investors appear increasingly willing to fund.

Market Context

The market focus is telling. Media and entertainment, Physical AI, and Robotics are 3 sectors where latency isn't a minor inconvenience. It's often the difference between a usable experience and one that never reaches production.

This funding also reflects a broader trend emerging across the AI ecosystem. Venture capital firms are increasingly backing companies that enable AI adoption rather than companies building foundation models themselves. As model capabilities improve, the infrastructure required to deploy, manage, and integrate those capabilities becomes increasingly valuable. Reactor sits within a growing category of AI infrastructure companies focused on making advanced AI systems practical for developers and enterprises.

What This Signals

A lesson sits beneath this round. Capital didn't flow because generative video is attracting attention. Capital flowed toward a team with deep technical experience that identified a missing layer in the ecosystem and built specifically for developers. That pattern appears repeatedly across technology cycles. Novel products attract attention, but infrastructure platforms often create lasting value because entire categories depend on them.

Lightspeed Venture Partners and the broader investor group are making a calculated bet that real-time AI experiences become a foundational layer of modern software. Reactor is making a parallel bet that developers will need better tools to build them. If world models become a meaningful part of how applications are created and experienced, $59M may be remembered less as a funding milestone and more as an early indicator of where the market was heading next.