Decart, an Israeli AI lab founded in 2023, has developed inference optimization technology that makes real-time generative video and interactive world models practical for the first time. The company's stack reduces GPU costs for video generation by orders of magnitude, from hundreds of dollars per hour to mere cents, while enabling sub-frame latency performance that powers live, interactive experiences.

Why Is Real-Time AI Generation So Hard?

The generative AI market is projected to grow from $103.6 billion in 2025 to $1.3 trillion by 2034, with generative AI in gaming alone expected to reach over $5.1 billion by 2030, up from $1.8 billion in 2025. Yet despite this explosive growth, most AI-generated content remains static, built for one-shot output rather than live, continuous interaction. The technical bottleneck is severe: diffusion models typically require many seconds of computation to produce a single second of video, while interactive applications demand generation in under 16 milliseconds per frame. Inefficiencies in how hardware is used mean real-time generation has remained largely out of reach.

This gap exists precisely where consumer demand is strongest. The global video game industry generated $189 billion in 2025 and served more than 3.6 billion players, while social and streaming platforms like TikTok and Twitch see billions of hours of user engagement each month. Consumers are spending increasing time in interactive, media-rich environments and are looking for tools that let them co-create in real time. Yet the infrastructure to enable that growing demand has been missing.

How Does Decart's Technology Work?

Decart targets this gap directly with a core innovation: an inference engine that drastically improves GPU utilization for generative media models. As of August 2026, Decart's stack has three layers: DOS, an inference-optimization layer sold to enterprises; Oasis, a world model that generates interactive environments; and Lucy, a live video model that edits streams as they run. All three share the same engineering premise: custom low-level work on how models run yields order-of-magnitude efficiency gains.

The results are striking. Decart claims its world models run 10 times more efficiently than alternatives, and the stack cuts the cost of GPU-based video generation from hundreds of dollars per hour to cents. These systems demonstrate that high-fidelity, responsive AI experiences are possible when software and hardware are optimized together. Mirage, one of Decart's products, enables live video-to-video transformation, while Oasis is a real-time AI-rendered game world that runs without a traditional game engine.

Who Built Decart and Why?

Decart was founded in September 2023 by Dean Leitersdorf, CEO; Moshe Shalev, Chief Product Officer; and Orian Leitersdorf, Chief Scientist. The founders brought complementary expertise: Dean Leitersdorf is a Technion-trained computer scientist who completed his BSc, MSc, and PhD by age 23, earning an ACM dissertation award. Moshe Shalev spent 13 years in Unit 8200, Israel's elite tech-intelligence corps, where he built and led large AI infrastructure projects.

The two met in Israel's intelligence community in mid-2021 and quickly recognized their complementary strengths. "We're both strong executors, but we dream differently," Shalev said of their dynamic, explaining that Decart is "the meeting point between Dean's ability to imagine technology on a broad scale and [my] ability to bring it into an organization and operationalize it". From the beginning, they set an unusually ambitious vision: to solve a truly massive problem and create an AI company as transformative as Google or Facebook.

What Challenges Did Decart Face Early On?

When Decart raised its seed round in mid-2024, the founders encountered skepticism. About half the investors they spoke with were excited and ready to invest, while the other half praised the technology but said they didn't understand the end product. One investor, an anonymous female investor nicknamed "Sarah," openly doubted the venture, saying Israelis "don't know how to train models" and have missed the generative AI wave.

Rather than being discouraged, Leitersdorf and Shalev took this as fuel, hanging a photo of "Sarah" on their office wall as a constant reminder to prove her wrong. The startup emerged from stealth with both an enterprise product and a consumer demo, and it began generating revenue immediately. The founders later summed up their mindset by saying they could have built something to sell off quickly, but instead they were determined to "build an app for a billion users," underscoring their long-term, ambitious outlook.

How to Understand Decart's Market Position

• Enterprise Focus: DOS, Decart's inference-optimization layer, targets enterprises seeking to reduce GPU costs and improve performance for generative media applications at scale.
• Consumer Experience: Oasis and Lucy demonstrate consumer-facing applications, from AI-rendered game worlds to live video editing, showing the practical impact of the technology on end users.
• Cost Efficiency: By cutting GPU costs from hundreds of dollars per hour to cents, Decart makes real-time generative experiences economically viable for platforms serving millions of concurrent users.
• Speed and Responsiveness: Sub-frame latency performance enables interactive experiences that feel natural and responsive, a critical requirement for gaming and live streaming applications.

As of 2026, Decart operates from Tel Aviv, San Francisco, and New York, with Kfir Aberman, a founding member who previously worked as a senior research scientist at Google and a principal research scientist at Snap, leading the San Francisco R&D center and US hiring. The company's stated mission is to make AI "fast, responsive and affordable enough to power dynamic, interactive experiences for millions of concurrent users".

The broader context matters here: as the generative AI market moves beyond static tools toward live environments, Decart is building the infrastructure needed to support that transition at scale. The company's technology addresses a fundamental bottleneck that has prevented real-time generative experiences from becoming mainstream. By solving the hardware efficiency problem, Decart is enabling a new category of interactive entertainment and content creation tools that were previously impossible to build economically.