U.S. Startup Fauna Launches Sprout: A Developer-Focused Humanoid Robotics Platform

January 27, U.S. Startup Fauna Robotics has unveiled Sprout, its first humanoid robot. Unlike the prevailing industry approach of prioritizing deployment in factories and warehouses, Sprout is designed from the ground up to operate in human-centric environments such as schools, exhibition spaces, and public areas. The initial production run, designated as the Creator Edition, is available for purchase by researchers, educators, and commercial developers.

 Source：Fauna Robotics

Features of Sprout Humanoid Robot

Compact Size and Approachable

Standing approximately 1.07 meters tall and weighing 22.7 kilograms, Sprout represents a marked departure from mainstream industrial humanoid platforms in both size and form factor. The system features extensive soft-bodied enclosures, combined with a low center of gravity and compliant control architecture that emphasizes yielding rather than resisting upon contact. 

The robot integrates time-of-flight (ToF) sensors for continuous environmental perception and obstacle avoidance. Fauna characterizes this comprehensive design philosophy as "safety-first for human spaces," targeting not enclosed or highly structured industrial settings, but rather unconfined coexistence with humans without requiring safety barriers.

Rob Cochran, Founder and CEO of Fauna (formerly of CTRL-labs, acquired by Facebook in 2019), notes that most existing humanoid systems remain fundamentally extensions of industrial machinery—large in mass, high in inertia, and poorly suited for close human-robot interaction. He argues that for robots to genuinely enter everyday environments, "approachability" and "acceptability" must be reconsidered at the design stage. Sprout deliberately softens industrial aesthetics through its green exterior, LED facial display, and articulated eyebrow mechanism, positioning it as an "interactive object" rather than conventional machinery.

Fauna Robotics is headquartered in New York City's Flatiron District, with manufacturing operations in the United States and engineering and production teams based in New York. The company currently employs approximately 50 people. Early adopters include Disney, Boston Dynamics, UC San Diego, and New York University, all exploring applications of the platform across retail, entertainment, home services, and related sectors.

Source：Fauna Robotics

Developers as the Target Users

Despite Sprout's form factor and operational context being closer to everyday human environments, Fauna has not positioned it as a consumer-ready product. Priced at approximately $50,000, the robot explicitly targets universities, research institutions, and corporate R&D teams. Its core positioning is as a "developer-first" humanoid platform, rather than a feature-complete, market-ready end product.

Developers can Customize Training Experiments

On the software side, Sprout provides a complete SDK and modular AI architecture, with built-in pre-trained locomotion capabilities including walking, kneeling, crawling, and sitting. Developers can integrate proprietary models at various system levels for experimentation in perception, control, or interaction logic. 

What is the Purpose of Fauna's Development of Humanoid Robots for Developers?

Josh Merel, CTO of Fauna (former DeepMind researcher), states that the company aims to lower barriers to humanoid application development, enabling a broader community of creators to rapidly prototype without requiring full robotics teams.

This platform-oriented approach parallels the developmental trajectory of early personal computers or smartphones, yet the practical barriers to robotics applications remain considerably higher. Unlike purely software ecosystems, humanoid robots must operate in physical environments, requiring hardware adaptation, safety validation, and extended testing cycles. The industry has yet to establish scalable, replicable application models.

From an industry perspective, humanoid robots remain predominantly deployed in industrial and research contexts, with systems capable of operating in shared human spaces still in nascent stages. 

Sprout's positioning does not refute this reality but rather seeks to provide, at an earlier inflection point, a repeatable validation platform for how humanoid robots can safely and controllably enter non-industrial environments. Its long-term value will depend on whether these early adopters can progressively construct application scenarios with tangible real-world utility.