Figure 01 - Robot Details, Use Case and Specifications

Figure 01

Figure 01 is a humanoid robot designed for manual labor, featuring advanced AI for autonomous operation and human-like interaction in industrial and logistics environments.

Enquire about robot



Data verified by manufacturer

Made by

Figure AI Inc.

Software Type

Closed Source

Software Package

Proprietary Figure AI locomotion control and motion planning stack. VLM developed in collaboration with OpenAI for real-time environmental reasoning. LLM integration for natural language command understanding and task sequencing. Cloud-based OTA update and remote diagnostics platform.

...

Report

Actuators

Custom electromechanical actuators provide smooth, human-like joint movements with high precision and strength, enabling dexterous manipulation and dynamic locomotion.

Compiute

Equipped with custom compute boards optimized for AI inference, sensor fusion, and real-time control, supporting complex decision-making and adaptive behaviors.

Sensors

Includes multi-modal sensors such as RGB cameras, LiDAR, and tactile sensors for comprehensive 3D perception, object recognition, and environmental awareness.

Max Op. time

300

mins

Recent Robot Videos

*Aparobot claims no ownership of videos posted unless otherwise stated.

Description text

Date

https://www.youtube.com/watch?v=K1TrbI0BaaU, Figure Status Update - BMW Full Use Case, 2024-07-02
https://www.youtube.com/watch?v=Sq1QZB5baNw&t=1sm, Figure Status Update - OpenAI Speech-to-Speech Reasoning, 2024-03-13
https://www.youtube.com/watch?v=Q5MKo7Idsok, Figure Status Update - AI Trained Coffee Demo, 2024-01-07
https://www.youtube.com/watch?v=-4erYt2t7Bs, Figure 01 Dynamic Walking, 2023-10-17
https://www.youtube.com/watch?v=jACJruCzUzY, Figure Status Update - Dynamic Walking, 2023-10-17

Robot Brief

Figure 01 is the first-generation bipedal humanoid robot developed by Figure AI in collaboration with OpenAI, introduced as a prototype in 2022 and taking its first steps in May 2023. Designed as the company's foundational proof-of-concept for general-purpose humanoid robots capable of performing manual labour in commercial environments, Figure 01 stands 1,680 mm tall and weighs 60 kg. It was the first humanoid robot to demonstrate real-time conversational reasoning and autonomous task planning in collaboration with OpenAI, and served as the primary platform through which Figure AI validated core capabilities including bipedal locomotion, dexterous manipulation, object recognition, and natural language task execution in real industrial settings. As of 2025, Figure 01 has been superseded by Figure 02 as Figure AI's active commercial platform and transitioned to legacy status.

Use Cases

Autonomous Task Execution in Industrial Environments:
Performs structured manual labour tasks including lifting and transporting objects up to 20 kg, operating machinery, opening doors, and climbing stairs without human supervision.
Natural Language Command Understanding:
Receives and interprets spoken task instructions through an onboard language model developed in collaboration with OpenAI, converting voice commands into coordinated physical action sequences.
Object Recognition and Contextual Interaction:
Identifies objects not only by shape but by function using visual language models, enabling appropriate interaction with tools, containers, and machinery across varied task scenarios.
Real-Time Error Correction and Task Adaptation:
Detects task execution failures during operation and self-corrects in real time using AI-driven feedback loops, without requiring human intervention to restart or redirect.
Environmental Navigation:
Traverses human-scale environments including corridors, stairways, and workspaces using cameras, LiDAR, and tactile sensors to perceive, map, and navigate its surroundings autonomously.

Industries

Logistics and Warehousing:
Developed primarily to address labour shortages in warehousing and distribution, with capabilities for material transport, object handling, and interaction with standard logistics infrastructure.
Manufacturing:
Validated in early commercial deployments as a platform for repetitive manual assembly and machine operation tasks in automotive and industrial manufacturing environments.

Specifications

Length

Width

Height (ResT)

Height (Stand)

1680

Height (Min)

Height (Max)

1680

Weight (With Batt.)

Weight (NO Batt.)

Max Step Height

Max Slope

+/-

Op. Temp (min)

°C

Op. Temp (Max)

°C

Ingress Rating

No items found.

Intro

Figure 01 stands 1,680 mm tall and weighs 60 kg. Its 41 degrees of freedom are distributed as 7 DoF per arm, 6 DoF per leg, and additional DoF in the hands, head, and torso, enabling a full human-scale range of motion across locomotion and manipulation tasks. Each hand provides four fingers and an opposable thumb, closely mirroring human hand geometry. Joints are actuated by custom electromechanical actuators — a rotary motor-based system designed in-house by Figure AI to convert electrical energy into mechanical force with the precision required for dexterous manipulation. The initial design featured externally routed cabling for ease of testing and modification, which was later integrated internally in Figure 02. The Figure 01 can carry payloads of up to 20 kg and walks at a maximum speed of 1.2 m/s.

Perception is provided by a combination of RGB cameras, LiDAR, and tactile sensors, which together enable 3D environmental mapping, object recognition, and physical interaction feedback. Onboard AI was developed in collaboration with OpenAI and includes a visual language model (VLM) for perceiving and reasoning about the environment, as well as a large language model for natural language command understanding and multi-step task planning. All inference runs onboard without cloud dependency for core task execution. A lithium-ion battery provides approximately 5 hours of continuous operation, and Figure 01 demonstrated operations in commercial industrial settings including initial BMW partnership pilots in 2024. The OpenAI collaboration concluded in early 2025 when Figure AI pivoted to developing its Helix VLA model entirely in-house.

Connectivity

Integration with OpenAI's language models for real-time processing of visual and auditory data.
Autonomous operation through neural networks enabling low-level motor actions and high-level reasoning.

Capabilities

Visual Language Model Perception:
An onboard VLM developed with OpenAI enables Figure 01 to understand its visual environment and reason about object properties, states, and appropriate interactions in real time.
Natural Language Task Planning:
Receives spoken commands and executes multi-step task sequences autonomously, linking language understanding to physical motor action through the onboard AI pipeline.
41-DoF Full-Body Articulation:
Seven DoF per arm, six DoF per leg, and hand and head articulation provide a complete human-scale range of motion across locomotion and manipulation.
Real-Time Error Detection and Correction:
Detects task failures mid-execution and self-corrects without human intervention, using closed-loop AI feedback to resume or modify task strategies.
Multi-Modal Environmental Sensing:
Combines RGB cameras, LiDAR, and tactile feedback to build a comprehensive real-time understanding of its environment, enabling object recognition, spatial mapping, and safe physical interaction.
Dynamic Bipedal Locomotion:
Walks at up to 1.2 m/s, climbs stairs, navigates doorways, and maintains balance across varied indoor terrains using real-time joint torque regulation.