Flexion Robotics Raises $50M Series A to Power AI ‘Brains’ for Humanoid Robots
Flexion Robotics, a Zurich-based startup building the intelligence layer for humanoid robots, has secured $50 million in Series A funding as it positions itself at the forefront of autonomous robotic intelligence. The company’s mission is to develop a reinforcement learning-driven autonomy stack that enables humanoid robots to perceive, reason, and act in dynamic real-world environments with minimal human intervention. The funding marks a major milestone for Flexion’s growth and underlines increasing investor interest in robotics and AI-driven autonomy solutions.
The Series A round was led by DST Global Partners, with participation from a group of notable venture investors including NVentures (the venture capital arm of NVIDIA), redalpine, Prosus Ventures, and Moonfire Ventures. The capital will support the company’s efforts to scale its research and development capabilities in Zurich, expand its compute infrastructure and robotic fleets, establish a U.S. presence, and accelerate the commercialization of its autonomy stack with major hardware partners.
Founded in 2024 by a team of engineers and scientists with backgrounds spanning ETH Zurich, NVIDIA, Meta, Google, Tesla, and Amazon, Flexion is focused on solving one of the toughest challenges in robotics: creating an adaptable, general-purpose software “brain” that can work across different robot bodies and application domains. Rather than building physical robots, the company’s approach is to provide a modular intelligence stack that OEMs and hardware manufacturers can integrate into their humanoid platforms. This software-first strategy sets Flexion apart from many competitors that develop integrated hardware and software products.
The newly raised funds will enable Flexion to accelerate delivery of its core autonomy platform, which combines a language-driven command layer, vision-language-action models trained largely in simulation, and transformer-based full-body control. This combination allows robots to understand natural language tasks, ground them in real environments, and generate complex motor actions — all without heavy reliance on teleoperation or brittle scripting. By leveraging synthetic training data and reinforcement learning in virtual environments, Flexion aims to reduce the cost and time required for robots to learn robust behaviors.
The company’s technology is already being tested with major original equipment manufacturers (OEMs), and the additional resources are expected to deepen these partnerships while propelling the platform toward commercial deployments. Flexion’s investors have highlighted the company’s potential to drive a new wave of robotic capability that can adapt to unpredictable settings such as manufacturing floors, logistics hubs, and hazardous environments where human presence is limited or costly.
Investor interest in Flexion reflects a wider surge of capital flowing into robotics and autonomous systems, particularly those leveraging artificial intelligence. The involvement of DST Global Partners signals confidence in the broader opportunity for AI-driven robotics intelligence. NVentures’ participation brings strategic alignment with NVIDIA’s deep commitments to AI and accelerated computing, while redalpine, Prosus Ventures, and Moonfire Ventures contribute venture expertise and access to networks that can support Flexion’s international growth.
In addition to Series A funding, Flexion previously raised a $7.35 million seed round earlier in 2025, backed by investors including Frst, Moonfire Ventures, and redalpine. That early backing helped the team advance its foundational research and begin building out its autonomy stack. The cumulative funding now positions the startup to accelerate its transition from research phase to commercial scale, as robotics companies worldwide compete to deliver practical, generalizable autonomy solutions.
Flexion’s leadership has emphasized that the company’s technology is designed to be hardware-agnostic, meaning it can run on a variety of humanoid robot platforms rather than being tied to a single body type. This flexibility could enable broader adoption across different use cases and hardware ecosystems, a strategy that differentiates Flexion from competitors focused solely on integrated hardware solutions.
As the global robotics market continues to evolve, Flexion Robotics’ $50 million funding round represents one of the more significant investments in autonomous intelligence platforms. The company’s focus on simulation-first reinforcement learning and modular autonomy positions it to play a central role in advancing how robots learn and operate. With these funds, Flexion is poised to strengthen its technical capabilities, expand its geographic footprint, and foster the commercial relationships needed to bring sophisticated humanoid intelligence to real-world applications.
