Commercial model
Pricing not public
Not publicly disclosed. That usually means the final commercial package depends on deployment scope, services, or negotiated terms.
Robot dossier
GR-3
Release
Aug 1, 2025
Price
Price TBA
Connectivity
2
Status
Active
Height
165cm
Weight
71kg
Battery
~2 hours (hot-swappable)
Payload
5 kg per arm
Fourier's third-generation humanoid and its first purpose-built "Care-bot," unveiled in August 2025. The GR-3 pivots the GRx platform away from research-only use toward healthcare, eldercare, and companionship. It retains the 55 degrees of freedom and 12-DoF dexterous hands of its predecessor but introduces an innovative soft-shell exterior with automotive-grade upholstery and 31 distributed pressure sensors for safe physical interaction with people. At its core is Fourier's proprietary Full-Perception Multimodal Interaction System, which fuses vision, audio, and tactile feedback into a real-time emotional processing engine. A dual-path decision architecture combines fast reflexive control with large-language-model reasoning for contextual dialogue. The robot demonstrated chess matches, dance routines, and natural conversation at CES 2026 in its first overseas appearance. Standing 165 cm tall and weighing 71 kg, the GR-3 supports hot-swappable batteries for around-the-clock operation and is ROS-compatible.
Listed price
Price TBA
Not publicly disclosed
Release window
Aug 1, 2025
Current status
Active
Fourier
Last verified
Apr 23, 2026
Share this robot
Open a plain share composer on X or Bluesky for this robot profile.
Technical overview
A fast read on the mechanical profile, sensing package, and platform integrations behind GR-3.
Height
165cm
Weight
71kg
Battery Life
~2 hours (hot-swappable)
Charging Time
Not disclosed (hot-swappable battery system)
Max Speed
Not disclosed
Payload
5 kg per arm
Operational profile
Capabilities
8
Connectivity
2
Key capabilities
Ecosystem fit
Explore further
Benchmark set
Shortcuts to the closest alternatives in the current ui44 set.
Humanoid
GR-1
Fourier
Price TBA
Humanoid
GR-2
Fourier
Price TBA
Humanoid
A2 Ultra
AGIBOT
Price TBA
Humanoid
Forerunner K1
Shanghai Kepler Exploration Robot Co., Ltd.
$30,000
The GR-3 is a Humanoid robot built by Fourier. Fourier's third-generation humanoid and its first purpose-built "Care-bot," unveiled in August 2025. The GR-3 pivots the GRx platform away from research-only use toward healthcare, eldercare, and companionship. It retains the 55 degrees of freedom and 12-DoF dexterous hands of its predecessor but introduces an innovative soft-shell exterior with automotive-grade upholstery and 31 distributed pressure sensors for safe physical interaction with people. At its core is Fourier's proprietary Full-Perception Multimodal Interaction System, which fuses vision, audio, and tactile feedback into a real-time emotional processing engine. A dual-path decision architecture combines fast reflexive control with large-language-model reasoning for contextual dialogue. The robot demonstrated chess matches, dance routines, and natural conversation at CES 2026 in its first overseas appearance. Standing 165 cm tall and weighing 71 kg, the GR-3 supports hot-swappable batteries for around-the-clock operation and is ROS-compatible.
Pricing has not been publicly disclosed. See all Fourier robots on the Fourier page.
Detailed specifications for the GR-3
Height
165cmAt 165cm, the GR-3 is designed to operate in human-scale environments, allowing it to reach countertops, shelves, and interfaces designed for human height.
Weight
71kgWeighing 71kg, the GR-3 needs to balance mass for stability during bipedal locomotion while remaining light enough for safe human interaction.
Battery Life
~2 hours (hot-swappable)With a battery life of ~2 hours (hot-swappable), the GR-3 can operate for sustained periods before requiring a recharge. Battery life is measured under typical operating conditions and may vary based on workload intensity and environmental factors.
Charging Time
Not disclosed (hot-swappable battery system)A charging time of Not disclosed (hot-swappable battery system) means the ratio of operation to downtime is an important consideration for applications requiring near-continuous availability. Some deployments use multiple robots in rotation to maintain uninterrupted service.
Payload Capacity
5 kg per armA payload capacity of 5 kg per arm determines what the robot can carry or manipulate. This is a critical spec for manipulation tasks, determining what objects the robot can lift, carry, and work with.
AI Platform
Fourier Full-Perception Multimodal Interaction System (dual-path: rule-based + LLM reasoning)The GR-3 uses Fourier Full-Perception Multimodal Interaction System (dual-path: rule-based + LLM reasoning) as its intelligence backbone. This AI platform powers the robot's decision-making, perception processing, and autonomous behavior. The sophistication of the AI stack directly impacts how well the robot handles unexpected situations and adapts to new environments.
The GR-3 integrates 6 sensor types, forming the perceptual foundation that enables autonomous operation.
This sensor configuration enables the GR-3 to perceive its 3D environment, recognize objects and people, navigate complex spaces, and perform precise manipulation tasks. Multiple sensor modalities provide redundancy and more robust perception than any single sensor type alone.
Explore sensor technologies: components glossary · full components directory
Humanoid robots are designed for environments built for humans — warehouses, factories, healthcare facilities, and eventually homes. Their bipedal form allows them to navigate stairs, doorways, and workspaces designed for human bodies without requiring environmental modifications.
The GR-3 offers 8 distinct capabilities, each contributing to the robot's practical utility.
These capabilities work together with the robot's 6 onboard sensor types and Fourier Full-Perception Multimodal Interaction System (dual-path: rule-based + LLM reasoning) AI platform to deliver practical, real-world performance.
The GR-3 integrates with the following platforms and ecosystems, extending its utility beyond standalone operation.
This ecosystem compatibility enables the GR-3 to work as part of a broader automation setup rather than operating in isolation.
8
Capabilities
6
Sensor Types
AI
Fourier Full-Perception Mult…
How the GR-3 communicates with your network, smart home devices, cloud services, and companion apps.
The GR-3 by Fourier integrates 9 distinct technology components across sensing, connectivity, intelligence, and interaction layers. The physical platform features a height of 165cm, a weight of 71kg, providing the foundation on which this technology stack operates.
The perception layer is built on RGB Camera, Structured-Light Depth Camera, 4-Microphone Array (voice localization, echo cancellation), 31 Distributed Pressure Sensors (full-body tactile), IMU, Force/Torque Sensors. These work in concert to give the robot a detailed understanding of its operating environment. This multi-sensor approach provides redundancy and enables the robot to function reliably even when individual sensors encounter challenging conditions such as low light, reflective surfaces, or cluttered spaces.
Fourier Full-Perception Multimodal Interaction System (dual-path: rule-based + LLM reasoning) serves as the computational brain, processing sensor data, making navigation decisions, and orchestrating the robot's autonomous behaviors. The quality of this AI platform directly influences how well the robot handles novel situations, adapts to changes in its environment, and improves its performance over time through learning.
Humanoid robots are typically targeted at enterprise customers, research institutions, and forward-thinking businesses looking to automate tasks that require human-like form and dexterity. While some models are approaching consumer pricing, the majority remain in the commercial and industrial space.
When evaluating a humanoid robot, payload capacity, degrees of freedom, and manipulation dexterity are critical factors. Battery life and charging time determine operational uptime. The AI platform determines how well the robot can adapt to new tasks and environments. Consider whether the robot needs to work alongside humans (requiring safety certifications) or will operate independently.
Pricing
The GR-3 has a status of Active. Check with Fourier for the latest availability details.
Engineering compromises and where this humanoid robot excels
With 6 sensor types onboard, the GR-3 has one of the more comprehensive perception systems in the humanoid category. This multi-modal approach enables robust environmental awareness, redundant obstacle detection, and reliable autonomous operation even in challenging conditions. More sensor diversity generally translates to better real-world adaptability.
With 8 distinct capabilities, the GR-3 is designed as a versatile platform rather than a single-task device. This breadth means the robot can handle varied scenarios and workflows, reducing the need for multiple specialized robots and increasing its utility across different situations.
With a payload capacity of 5 kg per arm, the GR-3 can handle meaningful physical tasks. This capacity enables practical applications like carrying tools, transporting materials, or supporting equipment mounts that lighter robots simply cannot accommodate.
At 71kg, the GR-3 is a substantial piece of equipment. This weight contributes to stability and robustness but also means the robot requires careful consideration of floor load limits, transportation logistics, and the potential impact force in the event of unexpected contact with people or objects.
Fourier has not published a public price for the GR-3. While common for enterprise-class robotics, the absence of transparent pricing can complicate budgeting and comparison shopping. Prospective buyers will need to engage directly with the manufacturer for quotes, which may vary by configuration and volume.
Note: This strengths and trade-offs assessment is based on the GR-3's documented specifications as tracked in the ui44 database. Real-world performance depends on deployment conditions, firmware maturity, and environmental factors. For the most current information, check the Fourier manufacturer page or visit the official product page. Use the comparison tool to evaluate these trade-offs against competing robots in the same category.
Understanding the engineering behind this category
Humanoid robots represent one of the most technically ambitious categories in robotics. Building a machine that walks, balances, manipulates objects, and interacts naturally with humans requires breakthroughs across multiple engineering disciplines simultaneously. Understanding the technology behind humanoid robots helps buyers and enthusiasts appreciate both the capabilities and limitations of current systems.
Humanoid robots navigate using a combination of visual SLAM (Simultaneous Localization and Mapping), depth sensing, and inertial measurement. Unlike wheeled robots that simply avoid obstacles, humanoids must plan footstep placement, maintain dynamic balance on uneven surfaces, and anticipate terrain changes. Advanced systems use predictive models to plan several steps ahead, similar to how humans unconsciously adjust their gait when approaching stairs or rough ground. The computational requirements for real-time bipedal navigation are substantial, often requiring dedicated motion-planning processors separate from the main AI system.
Artificial intelligence in humanoid robots serves multiple roles: high-level task planning (understanding what needs to be done), perception (recognizing objects, people, and environments), manipulation planning (figuring out how to grasp and move objects), and social interaction (understanding speech, gestures, and context). Modern humanoids increasingly use large language models and vision-language models for task understanding, allowing them to interpret natural language instructions and generalize to new tasks without explicit programming for each scenario.
The sensor suite in a humanoid robot must provide comprehensive environmental awareness while maintaining real-time processing speeds. Sensor fusion algorithms combine data from cameras, LiDAR, depth sensors, force/torque sensors, and IMUs to create a unified model of the robot's surroundings. This multi-modal perception is critical because no single sensor type works perfectly in all conditions — cameras struggle in darkness, LiDAR cannot distinguish materials, and touch sensors only detect what the robot physically contacts. By combining these inputs, the robot achieves more robust and reliable perception than any individual sensor could provide.
Battery technology is one of the primary limiting factors for humanoid robots. Bipedal locomotion is inherently energy-intensive — maintaining balance requires constant motor activity even when standing still. Current lithium-ion battery packs typically provide two to four hours of active operation, with charging times that can match or exceed operational time. Research into more efficient actuators, energy-harvesting techniques, and advanced battery chemistries aims to extend operational windows. Some commercial deployments address this limitation through battery-swap systems or scheduled charging rotations.
Safety in humanoid robotics is paramount because these robots operate in close proximity to humans. Design approaches include compliant actuators that absorb impact forces, real-time collision prediction systems, force-limited joints that automatically reduce power when unexpected contact occurs, and emergency stop mechanisms accessible to nearby humans. International safety standards like ISO 13482 for personal care robots provide frameworks for evaluating safety, but the field is still developing standards specific to general-purpose humanoid systems. Buyers should inquire about safety testing, certifications, and the robot's behavior in failure modes.
The humanoid robotics field is advancing rapidly on multiple fronts. Improvements in foundation models are enabling more generalizable intelligence. New actuator designs are making robots lighter and more efficient. Manufacturing scale is driving down costs. Over the next several years, expect humanoid robots to transition from controlled industrial environments to more varied commercial and eventually residential settings. The convergence of better AI, cheaper hardware, and proven deployment experience will accelerate adoption across industries.
The GR-3 by Fourier incorporates many of these technology pillars. For a detailed look at the specific sensors and components used in the GR-3, see the sensor analysis and connectivity sections above, or browse the complete components glossary for explanations of every technology used across the robotics industry.
How this robot compares in the humanoid landscape
Fourier has not publicly disclosed pricing for the GR-3, which is typical for enterprise-focused robotics platforms that offer customized solutions and direct-sales relationships.
With 6 sensor types, the GR-3 has an extensive sensor suite. This comprehensive sensing capability places it among the more perception-capable robots in the humanoid category, enabling more robust autonomous operation in varied conditions.
Being currently available for purchase gives the GR-3 a practical advantage over competitors still in development or prototype stages. Buyers can evaluate the actual product rather than relying on spec-sheet promises that may change before release.
Side-by-side specs, capability overlap analysis, and key differentiators.
For the full picture of Fourier's portfolio and market strategy, visit the Fourier manufacturer page.
What the public profile tells you, and what still needs direct vendor confirmation
From a buying and rollout perspective, the GR-3 should be read as a humanoid platform aimed at human-scale workplaces and pilot automation programs. ui44 currently tracks 8 capability signals, 6 sensor inputs, and a last verification date of 2026-04-23. That mix gives buyers a useful first-pass picture, but it is still only the public layer of due diligence, especially when procurement, uptime, and support commitments are decided directly with Fourier.
Commercial model
Pricing not public
Not publicly disclosed. That usually means the final commercial package depends on deployment scope, services, or negotiated terms.
Integration posture
2 connectivity options
The profile lists Wi-Fi, Ethernet, plus Fourier Full-Perception Multimodal Interaction System (dual-path: rule-based + LLM reasoning) as the AI stack. That is enough to infer the basic network posture, but buyers should still confirm APIs, fleet management, and workflow integration details. ui44 currently tracks 1 declared compatibility link.
Spec disclosure
5/7 core specs public
ui44 currently has 5 of 7 core physical and operating specs filled in for this model, leaving 2 gaps that matter for deployment planning. Missing runtime, charge, speed, or payload details can materially change staffing and site-readiness assumptions.
The current profile is detailed enough to support early comparison work, shortlist creation, and cross-checking against other humanoid robots. It is still worth validating the final deployment package, because integration services, support coverage, software entitlements, and site-preparation requirements often sit outside the raw hardware spec sheet.
If you want a faster apples-to-apples read, compare the GR-3 against nearby alternatives in ui44's compare view, then cross-check the underlying AI, sensor, and subsystem terms in the components glossary. For manufacturer-level context, the Fourier profile helps anchor this robot inside the wider product lineup.
Practical guide from day one through years of ownership
Setting up a humanoid robot is substantially more involved than plug-and-play consumer devices. Expect a professional installation or guided setup process that includes physical unpacking and assembly (if shipped disassembled), initial calibration of joints and sensors, environment mapping and safety zone definition, network and cloud service configuration, and application-specific programming or task teaching. Plan for several hours to a full day of setup time, and budget for potential integration consulting if the robot needs to connect with existing systems. The manufacturer or a certified integrator should provide training on safe operation, emergency procedures, and basic troubleshooting.
Humanoid robots require regular maintenance to ensure safe and reliable operation. Monthly maintenance typically includes visual inspection of joints and actuators for wear, sensor cleaning (especially cameras and LiDAR), firmware and software updates, battery health checks, and calibration verification. Quarterly maintenance may include more thorough mechanical inspection, lubrication of moving parts, and performance benchmarking to detect gradual degradation. Keep a maintenance log and follow the manufacturer's recommended schedule precisely — humanoid robots are complex systems where small issues can cascade if not addressed promptly.
Humanoid robot software is evolving rapidly, and regular updates can significantly improve performance, add new capabilities, and patch security vulnerabilities. Most manufacturers provide over-the-air updates, but enterprise deployments may require staging and testing updates before rolling them out. Evaluate the manufacturer's update track record — frequent, well-documented updates indicate active development and long-term commitment. Be aware that major software updates may require recalibration or retraining of custom behaviors.
To maximize the useful life of a humanoid robot, avoid operating beyond specified payload limits, maintain a controlled environment (temperature, humidity), keep sensors clean and unobstructed, and address any unusual sounds or behaviors promptly. Battery longevity is improved by avoiding deep discharges and extreme temperatures during charging. Investing in a service contract with the manufacturer or a certified partner provides access to replacement parts and expertise that can extend the robot's productive life significantly beyond the standard warranty period.
For Fourier-specific support resources and documentation, visit the Fourier page on ui44 or check the manufacturer's official website at Fourier's product page.
All GR-3 data on ui44 is verified against official Fourier sources, including spec sheets, product pages, and press releases. Last verified: 2026-04-23. Official source: Fourier product page. If you find outdated or incorrect information, please let us know — accuracy is our top priority.
See how the GR-3 stacks up — compare specs, browse the humanoid category, or search the full database.