KORTHOS / ECOSYSTEM / PRODUCT

Booster T1

About

About Booster T1

Booster T1 is an approximately 1.18-metre, 30-kilogram humanoid configurable from 23 to 41 degrees of freedom. Dual-encoder joints provide peak torque reaching 130 newton-metres, while optional seven-axis arms, grippers and dexterous hands extend the platform from locomotion research into manipulation.

NVIDIA Jetson AGX Orin, depth sensing and inertial measurement support onboard policies and perception. Open APIs, documentation, simulation tools and reinforcement-learning frameworks connect the robot to research pipelines. Battery endurance is approximately two hours while walking and four hours while standing. Booster has supplied T1 to university research groups and RoboCup teams.

17 APR 2026

Booster Robotics raises nearly 1 billion yuan as Q1 shipments surge 500%

Capabilities

Source-backed capabilities

  • Operating environments: IEEE-RAS Humanoids 2024 in Nancy
  • Bipedal walking

    Demonstrated - 2024-11-24 - Walk through the Humanoids 2024 speed-competition course. - environment: IEEE-RAS Humanoids 2024 in Nancy

    Source
  • Navigation and obstacle avoidance

    Demonstrated - 2024-11-24 - Avoid obstacles in the Humanoids 2024 challenge. - environment: IEEE-RAS Humanoids 2024 in Nancy

    Source
  • Multi-terrain locomotion

    Research demonstration - 2025-06-18 - The authors transferred trained policies to physical Booster T1 hardware for omnidirectional walking, disturbance resistance and terrain adaptation.

    Source
  • Doors and articulated objects

    Demonstrated - 2024-11-24 - Open a door in the Humanoids 2024 challenge. - environment: IEEE-RAS Humanoids 2024 in Nancy

    Source

Operations

  • Shipments

    more than 50 Booster T1 units shipped - 2025-01-23 - global Booster T1 customers - humanoid research and competition robotics - Booster Robotics said Booster T1 had reached mass production and global shipment, with more than 50 units shipped within three months of launch.

    Source

Supply

Supply / Components

  • Booster T1

    Jetson Agx Orin For Booster T1 - Compute & Control - NVIDIA - Booster T1 incorporates NVIDIA Jetson AGX Orin compute - source: booster.tech

    Source
  • Booster T1

    Core I7 1370p For Booster T1 - Compute & Control - Intel - Intel supplies core i7 1370p for booster t1 for Booster Robotics Booster T1 - source: booster.tech

    Source

Research

  • InterMimicGen: Scaling Humanoid Loco-Manipulation through Self-Evolving Motion Imitation

    2026-10-05 - Research Publication - InterMimicGen couples humanoid motion references with a learned tracker in an iterative data-generation loop. It retargets human-object interactions to several embodiments and demonstrates selected motions on physical G1, Booster K1 and Dexmate Vega-1 Pro robots.

    Source
  • Vision-Based Dribbling for Humanoid Soccer via Privileged Representation Learning

    2026-07-14 - Research Publication - The authors propose an integrated approach in which a temporal depth encoder is embedded into a reinforcement learning policy through a task-specific projection layer.

    Source
  • Reinforcement Learning-Based Control for an Inline Skating Humanoid Robot

    2026-06-30 - Research Publication - A Booster T1 fitted with commercial inline skates executes learned skating control.

    Source
  • Booster Lab: A Data-Centric Pipeline for Learning Deployable Humanoid Locomotion Policies

    2026-06-26 - Research Publication - The pipeline supports outdoor T1 locomotion and stable physical K1 running.

    Source
  • VOFA: Visual Object Goal Pushing with Force-Adaptive Control for Humanoids

    2026-05-02 - Research Publication - A physical Booster T1 pushes objects up to 17 kilograms with over 80% reported trial success.

    Source
  • FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control

    2026-04-06 - Research Publication - FlashSAC scales off-policy reinforcement learning with larger networks, high-throughput data and bounded optimization norms. It benchmarks diverse simulation tasks and transfers blind locomotion policies to a physical 29-DoF Unitree G1, including flat motion and stair climbing.

    Source
  • AGILE: A Comprehensive Workflow for Humanoid Loco-Manipulation Learning

    2026-03-20 - Research Publication - Recent advances in reinforcement learning (RL) have enabled impressive humanoid behaviors in simulation, yet transferring these results to new robots remains challenging.

    Source
  • REFINE-DP presents RL fine-tuning method for data-efficient humanoid loco-manipulation

    2026-03-18 - Research Publication - REFINE-DP jointly fine-tunes Diffusion Policy planner and RL whole-body controller for humanoid loco-manipulation, validated on Booster T1 with 90%+ success.

    Source
  • HybridMimic: Hybrid RL-Centroidal Control for Humanoid Motion Mimicking

    2026-03-06 - Research Publication - Motion mimicking, i. e.

    Source
  • Embodiment-Aware Generalist Specialist Distillation for Unified Humanoid Whole-Body Control

    2026-02-03 - Research Publication - Humanoid Whole-Body Controllers trained with reinforcement learning (RL) have recently achieved remarkable performance, yet many target a single robot embodiment.

    Source
  • Learning Sim-to-Real Humanoid Locomotion in 15 Minutes

    2025-12-01 - Research Publication - Massively parallel simulation has reduced reinforcement learning (RL) training time for robots from days to minutes.

    Source
  • AutoOdom evaluates autoregressive proprioceptive odometry on Booster T1

    2025-11-24 - Research Publication - Accurate proprioceptive odometry is fundamental for legged robot navigation in GPS-denied and visually degraded environments where conventional visual odometry systems fail.

    Source
  • PACE: Physics Augmentation for Coordinated End-to-end Reinforcement Learning toward Versatile Humanoid Table Tennis

    2025-09-25 - Research Publication - Humanoid table tennis (TT) demands rapid perception, proactive whole-body motion, and agile footwork under strict timing--capabilities that remain difficult for end-to-end control policies.

    Source
  • SEEC: Stable End-Effector Control with Model-Enhanced Residual Learning for Humanoid Loco-Manipulation

    2025-09-25 - Research Publication - This model covers SEEC: Stable End-Effector Control with Model-Enhanced Residual Learning for Humanoid Loco-Manipulation for Booster T1 in dexterity.

    Source
  • Booster Gym: An End-to-End Reinforcement Learning Framework for Humanoid Robot Locomotion

    2025-06-18 - Research Publication - Booster Gym releases an end-to-end reinforcement-learning framework spanning humanoid policy training, cross-simulator checks and physical deployment. The coauthored study validates locomotion transfer to Booster T1, and the official repository includes source and a pretrained T1 policy.

    Source
  • fan-ziqi/robot_lab

    2024-07-06 - Research Publication - The source registers exact model-specific Isaac Lab environments for tracked quadruped, wheeled-quadruped and humanoid platforms.

    Source
  • Booster RoboCup software stack

    Research Publication - Booster’s RoboCup release combines object detection, match decision-making and GameController integration for K1 and T1. The ROS2 stack includes behavior trees, vision configuration and hardware/simulator run scripts.

    Source

Resources

Media library

Official operating and developer references, exact-variant imagery and a verified instructional video for the Booster T1 humanoid platform.

Timeline

  • EAGLE study transfers control across four humanoids

    2026-02-03 - Research Publication - A policy trained in simulation controlled physical Unitree H1, Unitree G1, Booster T1 and Fourier N1 robots without hardware-specific fine-tuning. Hardware demonstrations covered walking, leaning and squatting; a fifth platform, PNDbotics Adam, was included in simulation.

    Source
  • Booster T1 powers Tsinghua’s RoboCup 2025 AdultSize title

    2025-07-21 - Award - The final in Brazil paired Tsinghua with China Agricultural University’s Mountain & Sea team, giving Chinese teams both top places. Tsinghua’s group-stage results included 16–0, 9–0 and 12–0 wins using the T1 humanoid platform.

    Source
  • Booster T1 serves as the platform for RoBoLeague finals

    2025-06-28 - Demonstration - Tsinghua Hephaestus won the finals ahead of China Agricultural University, Tsinghua’s Future Lab and Beijing Information Science and Technology University teams. Teams prepared their own AI strategies and robots made decisions on the field without handheld remote control. The tournament was a test event for the 2025 World Humanoid Robot Games.

    Source
  • Booster Robotics releases Booster Gym locomotion framework

    2025-06-18 - Research Publication - The open-source framework combines reinforcement learning, domain randomization, reward design and handling of parallel mechanical structures. The authors transferred trained policies to physical Booster T1 hardware for omnidirectional walking, disturbance resistance and terrain adaptation.

    Source
  • Booster T1 powers German Open AdultSize champions

    2025-04-14 - Award - The AdultSize final was played against Germany’s Sweaty team at the tournament in Nuremberg. Hamburg Bit-Bots also tested its own soccer software on a borrowed T1, retaining Booster’s motion control while adding localization and ball-blocking behavior.

    Source
  • Booster Robotics reports more than 50 T1 shipments

    2025-01-23 - Production Milestone - Customers included organizations in China, the United States, the United Arab Emirates, Germany, Switzerland, Japan and South Korea. The update accompanied T1’s January 19 appearance at NVIDIA’s Spring Festival reception, where it greeted guests and received Jensen Huang’s signature.

    Source
  • Booster T1 wins two Humanoids 2024 competitions

    2024-12-02 - Award - The competitions took place during IEEE-RAS Humanoids 2024 in Nancy, France, from November 22 to 24. Booster had launched T1 in October 2024, shortly before the November competitions.

    Source
  • Booster Robotics shows T1 at the 2024 World Robot Conference

    2024-08-21 - Demonstration - An August 21 visit included demonstrations of autonomous get-up and omnidirectional walking. The stand also showed ball throwing, kicking, push-ups and a low squatting movement, including interaction with visitors.

    Source

Context

Market context

Booster T1 is a higher-performance open humanoid for developers and researchers with ROS 2 integration, simulation interfaces and secondary development support.

  • Target environment: Research labs, university programs and developer workshops.
  • Workflow context: Humanoid development, locomotion research, embodied AI experimentation and secondary development using an openly programmable platform.
  • Customer context: Robotics researchers, AI labs and developers seeking a programmable humanoid platform.
  • Deployment model: Commercial product positioned as an open research and development platform.
  • Commercial maturity: Commercially available developer humanoid with ROS 2 and simulation support.
  • Adoption constraints: Designed for research and development rather than industrial deployment.
  • Market position: Sits between educational humanoids and industrial systems as an openly programmable research platform.
  • Adjacent products: G1, H1, AI Sapiens K0

Comparable Systems

  • AI Sapiens K0

    ROBOTIS - humanoid robotics

  • Booster T2

    Booster Robotics - humanoid robotics

  • G1

    Unitree - humanoid robotics

  • G1+

    Unitree - humanoid robotics

  • H1

    Unitree - humanoid robotics

  • H2

    Unitree - humanoid robotics

Related Systems

  • Digit

    Agility Robotics - humanoid robotics