Hugging Face released LeRobot Humanoid, a $2,500 bipedal robot platform built from 3D-printed parts and off-the-shelf components, targeting researchers testing AI-powered robotics software in physical experiments. The open-source release includes a bill of materials, 3D-printable part files, wiring documentation, physical assembly instructions, and software tools for calibration, control, and simulation. The design aims to bridge the gap between simulated robot training and real-world validation.
The platform prioritizes affordability and repairability over cutting-edge performance. LeRobot Humanoid uses printable parts, affordable actuators, and modular electronics to enable rapid iteration and modification. The team designed the robot for reproducible full-robot design loops where behaviors trained in simulation can be tested in physical bodies, with real-world trial data feeding back into simulation improvements.
Hugging Face CEO Clem Delangue is backing open-source robotics projects to keep development costs low and counter industry consolidation by large corporate players. In May 2025, the company partnered with French firm The Robot Studio to develop HopeJR, a humanoid robot with 66 actuated degrees of freedom targeting a $3,000 price. Hugging Face also sells Reachy Mini, a $299 robot designed for expressive interactions.
Key facts
- 01Hugging Face released LeRobot Humanoid, a 3D-printable bipedal robot platform starting at $2,500.
- 02Venture capital funding for robotics tripled between 2023 and 2025 to surpass $40 billion in 2025.
- 03Commercial humanoid robots typically cost $30,000 to $150,000 per unit, per McKinsey's April 2026 report.
- 04Unitree Robotics reported a 53% drop in first-quarter profits despite 68% revenue growth.
- 05Hugging Face previously partnered with The Robot Studio on HopeJR, a humanoid robot with 66 actuated degrees of freedom targeting a $3,000 price.
The robotics hardware market is splitting into two tracks. Commercial humanoid robots still cost $30,000 to $150,000 per unit as companies scale supply chains, according to an April 2026 McKinsey report. Venture capital funding for robotics tripled between 2023 and 2025 to surpass $40 billion last year. Chinese manufacturer Unitree Robotics already sells models below $20,000 and is pursuing a $610 million IPO on Shanghai's Star Market.
Unitree reported a 53% drop in first-quarter profits despite 68% revenue growth, a margin squeeze the South China Morning Post attributed to soaring expenses and a brutal price war. The company acknowledged a possible cooling of hype surrounding humanoid robots. Hyundai Motor Group is planning to mass-produce Boston Dynamics' Atlas humanoid robot at its Georgia electric vehicle plant and discussing a US-based facility capable of producing 350,000 robotic actuators annually.
LeRobot Humanoid ships with legs only. Hugging Face plans to integrate an upper body and develop more advanced behaviors. The company previously released a 3D-printable robotic arm. The team designed LeRobot Humanoid as a platform for iterative development rather than a one-off prototype.
The shift toward sub-$5,000 humanoid platforms matters because physical validation remains the bottleneck for AI robotics development. Simulation-only training produces models that fail in the real world; commercial robots priced above $30,000 keep experimentation limited to well-funded labs. A $2,500 bipedal robot with full design files and simulation tools lets smaller research teams and independent developers test hypotheses in hardware at scale.
Whether open-source hardware can compete with vertically integrated manufacturers depends on how quickly the developer community builds on Hugging Face's foundation. Boston Dynamics and Unitree control their supply chains and can optimize for volume manufacturing. Hugging Face is betting that a large base of contributors iterating on a shared platform will produce better software faster than any single company can. The LeRobot Humanoid release tests that thesis in the robotics hardware stack.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
