Web3 Robotics Research Report: The Current State and Outlook of Decentralized Machine Economies and Embodied Intelligence
TechFlow Selected TechFlow Selected
Web3 Robotics Research Report: The Current State and Outlook of Decentralized Machine Economies and Embodied Intelligence
Currently facing challenges in technical integration, interactive experience, user education, and other aspects, the paradigm shift brought by combining embodied intelligence with Web3 still holds tremendous potential.
Navigating Web3 tides with focused insightsExecutive Summary
This report explores how technological trends dominated by traditional tech giants migrate into the cryptocurrency market, predicting this cyclical pattern will repeat in the humanoid robotics sector.
The report first reviews the evolution paths of the metaverse and artificial intelligence waves, then analyzes global tech capital investment dynamics in embodied intelligence, and introduces progress and challenges of top humanoid robot companies.
Finally, it focuses on an overview of Web3 robotics and embodied intelligence projects, covering key project characteristics, economic models, and growth potential.
Through deep analysis, this report aims to provide industry participants with cutting-edge insights, revealing the strategic significance of decentralized machine economies in bridging the physical and digital worlds.
I. The Cyclical Pattern: Traditional Tech Giants Define Trends, Wall Street Capital Quickly Follows, Crypto Markets Efficiently Replicate as Decentralized Investment Opportunities—May Reappear in Humanoid Robotics
1. Retrospective: Migration of Mainstream Narratives from Tech Giants to Crypto
The narrative evolution in Web3 is deeply influenced by movements of traditional tech giants and financial capital. Reviewing past narrative cycles, the core trajectory can be summarized as: traditional tech giants define trends, Wall Street capital quickly follows, and crypto markets efficiently replicate them as decentralized investment opportunities.
Metaverse Wave (2021–2022):
-
In October 2021, Meta (formerly Facebook) led the rebranding and announced a full transition to the metaverse, investing tens of billions of dollars into VR/AR ecosystems. Microsoft launched its Mesh for Teams mixed-reality platform in November of the same year, focusing on enterprise virtual collaboration. NVIDIA simultaneously released its Omniverse digital twin platform, while Qualcomm established a $100 million fund and introduced Snapdragon Spaces to expand AR ecosystems. This sequence marked comprehensive internet giant布局 in virtual worlds.
-
Crypto assets also performed significantly, with metaverse platform tokens such as Decentraland (MANA) and The Sandbox (SAND) surging by 70x and 280x respectively. NFT floor prices for virtual lands like Decentraland LAND and The Sandbox LAND briefly reached around 5 ETH and 4 ETH, equivalent to tens of thousands of dollars. Top NFT projects like Bored Ape Yacht Club (BAYC) rose from an initial mint price of 0.08 ETH to over 150 ETH, exceeding $400,000 in value. Driven by celebrity effects, big-tech entry, and speculation, metaverse assets achieved unprecedented short-term explosive growth during this period. Although the market later corrected sharply, their astonishing gains remain iconic events in crypto history.
AI Wave (2023–2024):
-
In late 2022, OpenAI’s ChatGPT ignited a global AI boom, marking the dawn of generative AI. Microsoft responded swiftly, announcing a $10 billion investment in OpenAI in January 2023 and deeply integrating its technology into Bing search and the Office suite. Google followed with Bard conversational AI (later upgraded to Gemini) and restructured DeepMind to accelerate AI R&D. Meta adopted an open-source strategy, successively releasing Llama series large language models to advance ecosystem development.
-
Starting in 2023, multiple AI-themed tokens attracted significant funding, with platform tokens like Fetch.ai (FET) and Render Token (RENDER) showing notable price increases—surpassing 10x highs by early 2024. Decentralized AI projects like Bittensor (TAO), aiming to build decentralized neural network training and incentive mechanisms, delivered over 10x excess returns compared to mainstream assets. By late 2024, the surge in AI Agent-related infrastructure sparked broader market interest, generating several hundred-fold opportunities within a short time, fully igniting FOMO sentiment in the crypto market toward the AI sector.
2. Global Tech Capital Doubling Down on Embodied Intelligence: Humanoid Robots Emerge as the Next Narrative Focus
Currently, attention from traditional tech giants and top-tier VCs toward humanoid robots is rapidly intensifying. This chain reaction of “tech trend → hot capital → narrative migration” is repeating, shifting focus from AI and metaverse to embodied intelligence and humanoid robotics, positioning it as a leading candidate for the next Web3 narrative.
Morgan Stanley: Humanoid Robots Will Reshape the Global Labor Market
In February 2025, Morgan Stanley published a research report titled *Humanoid 100: Mapping the Humanoid Robot Value Chain*, forecasting that about 75% of U.S. jobs (63 million positions) possess "humanoid robot compatibility," potentially affecting up to $2.96 trillion in wage expenditures by 2050. Globally, the potential market size could reach $9 trillion.
The report divides the industrial chain into three core modules:
-
Brain: Includes foundational AI models (e.g., NVIDIA's Project Groot), data simulation and modeling (e.g., Palantir, Oracle), vision systems, and semiconductors (NVIDIA, Intel, Qualcomm).
-
Body: Composed of actuators (NSK, RBC Bearings), sensors (Analog Devices, Robosense), and battery systems (CATL, LG Energy), emphasizing lightweight design.
-
Integrators: Companies such as Tesla, Apple, Samsung, Xiaomi, Amazon, and Alibaba capable of building complete robotic systems.
Goldman Sachs: 2025 Could Be the Year of Mass Production for Humanoid Robots
In its February 2025 report *Humanoid Robots III*, Goldman Sachs noted that humanoid robots are entering a "supply chain leap phase," with Asian manufacturers rapidly joining and a nascent global industry ecosystem forming. The sector is projected to reach $38 billion annually by 2035, surpassing $200 billion under optimistic scenarios.
Companies such as 1x (Norway), Figure AI (USA), Agility Robotics (USA), Unitree (China), AGIBOT Zhiyuan (China), and Leju Robotics (China) have already launched commercial products. Others—including Tesla (USA), Sanctuary AI (Canada), Boston Dynamics (South Korea), Apptronik (USA), UBTECH (China), Kepler (China), Xpeng (China), Fourier Intelligence (China), Galaxy General (China), and Zhejiang Humanoid Robot Innovation Center (China)—are currently in factory trial production.
Tesla: Robots Will Become a "Trillion-Dollar" Growth Engine
On January 29, 2025, during Tesla’s Q4 2024 earnings call, CEO Elon Musk stated that humanoid robots would eventually become the mainstay of industry, possibly outnumbering humans with projections of 10 to 20 billion units. He aims to produce one billion units annually, capturing over 10% market share, creating a $25–30 trillion market opportunity for Tesla.
NVIDIA: The ChatGPT Moment for General-Purpose Robots Is Coming
At COMPUTEX 2025 on May 21, 2025, NVIDIA predicted that specialized robots would dominate adoption in the next 3–5 years (especially in manufacturing and logistics), though non-technical factors like safety certification and social acceptance may limit deployment. In roughly 10 years, general-purpose robots could become part of daily life, requiring breakthroughs in hardware clusters, simulation efficiency, data integration, and reliance on "scale laws."
NVIDIA CEO Jensen Huang announced Isaac GR00T N1 at the 2025 GTC conference—the first open foundational model platform for robotics, using an architecture mimicking human reflex and reasoning dual mechanisms. Simultaneously, NVIDIA launched Cosmos simulation platform and Newton physics engine to generate high-quality training data and accelerate robot interaction with the real world.
Additionally, NVIDIA’s Blue entertainment robot signals that robots are no longer confined to industrial settings but are entering consumer and interactive experience domains.
3. Overview of Top Global Humanoid Robot Companies
(1) Tesla Optimus
Tesla’s general-purpose humanoid robot Optimus targets factory tasks and home applications, relying on pure-vision AI and Full Self-Driving (FSD) technology, with plans for mass production in 2025 and a target price below $30,000.
Development Timeline
-
August 2021: First AI Day unveiled Optimus concept art, announcing the humanoid robot initiative.
-
October 2022: Initial bare-metal version showcased at AI Day.
-
March–December 2023: Achieved walking, object pickup, visual recognition, simple actions (e.g., yoga), and end-to-end neural network training.
-
February–October 2024: Gen 2 released with enhanced sensors and controllers, more stable gait, autonomous navigation, charging, stair climbing, and human interaction capabilities.
-
From 2025: Plans to test thousands internally; target 50,000–100,000 units in 2026, scaling to 500,000–1 million by 2027.
Technical Features
-
Mechanical Performance: Gen 2 stands ~1.73m tall, weighs 57kg, carries 20kg; Gen 3 features 22-degree-of-freedom dexterous hands.
-
Power System: Uses EV batteries and drive systems for long endurance and high energy efficiency.
-
Motion Capabilities: Supports walking, squatting, object pickup, single-leg standing, and other basic functions.
-
Perception & Control: Utilizes FSD computing platform for autonomous navigation and environmental perception.
-
Actuation Structure: 14 rotary joints (frameless motors + harmonic drives), 14 linear joints (planetary roller screws), totaling 70 bearings.
Key Achievements
-
Gen 2 prototype has demonstrated basic industrial operations and autonomous navigation.
-
FSD AI platform successfully ported to robotic system.
-
Pilot Deployment: 1,000 units deployed in Tesla factories starting 2024.
-
Market Plan: Thousands to be produced in 2025, priced at $20,000–30,000.
-
Long-Term Goal: Achieve billion-unit scale to replace repetitive manual labor.
(2) Figure AI
Founded in 2022 by Brett Adcock, Figure AI is a U.S.-based robotics company focused on developing AI-driven humanoid robots to address labor shortages, particularly in manufacturing and retail.
Development Timeline
-
May 2022: Company founded, focusing on humanoid robot R&D.
-
October 2023: Launched Figure 01 integrated with GPT-4, enabling natural language interaction and item classification.
-
February 2024: Closed $675M Series B round (investors include Microsoft, NVIDIA, OpenAI), valuing the company at $2.6B.
-
August 2024: Released Figure 02 with full hardware upgrade, 16-DOF dexterous hand, 20-hour battery life.
-
November 2024: Received demand for 40 units; targeting 2,000 units by April 2025, 20,000 units in Q3.
-
February 2025: Seeking $1.5B Series C round, valuation approaching $40B.
-
Future Plan: Ship 100,000 robots within four years; new production facility to launch in 2025.
Technical Features
-
Visual & Perception: Equipped with six RGB cameras and LiDAR, offering 360° environmental awareness.
-
Hand Capability: 16 degrees of freedom per hand, capable of lifting 25kg with high precision.
-
Computing Platform: Powered by NVIDIA RTX GPU, delivering triple the inference speed of previous generation.
-
AI System: Employs hierarchical embodied large model architecture (planning, decision-making, execution layers), reducing dependency on OpenAI through proprietary end-to-end AI development.
Key Achievements
-
Factory Testing: Completed 7×24 hour tests at BMW plants, performing assembly and transport tasks.
-
Commercial Progress: Secured supply agreements with major enterprises; targeting 2,000 units by April 2025, mass production of 20,000 units in Q3.
-
Funding: Raised $675M in Series B (2024), valued at $2.6B; seeking $1.5B Series C (2025), ~$40B valuation.
-
Future Outlook: Deploy 100,000 robots within four years across industrial and household scenarios.
(3) Unitree (Unitree)
Founded in 2016 in Hangzhou, China, Unitree is a high-tech enterprise specializing in high-performance legged and general-purpose humanoid robots. Known for fully self-developed core components and motion control algorithms, its product line spans consumer, industrial, and performance applications. Pursuing a "low-cost, high-performance" strategy, Unitree has driven robot commercialization forward—achieving over 60% global market share in quadruped robots and pioneering mass production of humanoid robots.
Development Timeline
-
2016: Founded in Hangzhou, founding team from Zhejiang University and Huawei, initially focused on quadruped locomotion algorithm R&D.
-
2017: Launched first consumer-grade quadruped robot Laikago (22kg weight, 5kg payload), gaining industry attention for being cheaper than Boston Dynamics' equivalents.
-
2019–2021: Released industrial Aliengo, educational A1, and companion robot Go1.
-
2022: 109 Go1 robot dogs appeared at Beijing Winter Olympics opening ceremony; launched industrial B1 and fitness pump PUMP.
-
2023: Released upgraded Go2 with 4D LiDAR; launched H1, China’s first full-size running humanoid robot, and industrial B2 quadruped.
-
2024: Closed nearly RMB 1B Series B2 round; launched humanoid robot Unitree G1; ranked in Forbes China AI Top 50.
-
2025: Humanoid robot H1 "Fuxi" performed dance routine “YangBOT” at CCTV Spring Festival Gala; G1 and H1 began limited sales at ¥99,000 / ¥650,000.
Technical Features
-
Full Supply Chain R&D: Covers motors, reducers, controllers, LiDAR, and high-performance motion control algorithms.
-
High-Performance Mobility: Go1 enables natural follow behavior; Go2 equipped with 4D ultra-wide-angle LiDAR; B2 industrial robot runs at 6m/s across complex terrain; H1 humanoid demonstrates dynamic balance and fast sprinting.
-
AI Enhancement: Trained via large models to achieve intelligent interactions like dance learning and music response.
-
Low-Cost Hardware: Through optimized motor drivers, reducer designs, and supply chains, reduced quadruped cost to one-tenth of industry average.
Key Achievements
-
Product Portfolio: Consumer (Go series), Industrial (B series), Humanoid (H1/G1).
-
Market Performance: Over 60% global market share in quadrupeds; Go series popular in STEM education. Pre-orders for G1 and H1 sold out quickly on JD.com, with 50% overseas orders.
-
Global Influence: Appeared at CES, Super Bowl pre-game show, Winter Olympics; products sold globally.
-
Technical Breakthroughs: BeamDojo reinforcement learning enables G1 to walk on梅花桩 and balance beams, resisting external disturbances; HOMIE cockpit system allows full-body remote control for complex tasks like dancing and carrying.
-
Commercialization: Limited sales of H1/G1 began in 2025, accelerating scalable application of general-purpose robots.
(4) Apptronik
Apptronik, headquartered in Austin, Texas, USA, was founded in 2016 and specializes in developing general-purpose humanoid robot Apollo, designed to enhance efficiency and safety in industries such as manufacturing, logistics, and healthcare through human-robot collaboration.
Development Timeline
-
2016: Spun out from University of Texas Human-Centered Robotics Lab, beginning dedicated robotics R&D.
-
2016–2019: Secured multiple government and private grants (including NASA collaborations), developing exoskeletons, humanoid upper bodies, bipedal platforms, and logistics arms.
-
2020: Developed humanoid upper body robot.
-
2022: Built fully electric humanoid prototype in under 12 weeks.
-
August 2023: Launched Apollo Alpha, marking the debut of the Apollo series.
Technical Features
-
Size & Payload: Apollo stands ~173cm, weighs ~73kg, capable of carrying ~25kg.
-
Power System: Features 7-DOF robotic arms with shoulder, elbow, and wrist joints for precise manipulation.
-
Sensors & Perception: Integrated depth camera and ultra-HD long-range LiDAR to enhance environmental perception in complex settings.
-
Safety Design: Unique force-control architecture ensures safe human collaboration, functioning more like collaborative robots than traditional industrial ones.
Key Achievements
-
Industrial Applications: Apollo tested at Mercedes-Benz plants in Marienfeld and Kecskemét, handling parts transportation and quality inspections.
-
AI Integration: Collaborates with NVIDIA and Google DeepMind to improve operational capabilities via AI.
-
Partnerships: Works with supply chain giant Jabil to manufacture and test Apollos in factories—even planning robots to assist in their own production.
-
Funding: Closed $403M Series A in February 2025, valued at ~$1.5B, with investors including Google DeepMind, B Capital, and Capital Factory.
-
Awards: Apollo nominated for Fast Company’s 2024 “World-Changing Ideas Award” in Experimental category, recognized for its potential in labor shortage and industrial automation.
-
Future Plans: Target hundreds of units in 2025, full commercialization in 2026, with goal of sub-$50,000 pricing.
(5) Boston Dynamics Atlas
Boston Dynamics, based in Massachusetts, USA, is renowned for developing highly mobile and agile humanoid robots used widely in industrial automation, logistics, and security.
Development Timeline
-
1992: Founded, initially conducting robotics research for the U.S. Army.
-
2005: Launched BigDog, a military quadruped robot (payload 154kg).
-
2013: Acquired by Google, later sold to SoftBank in 2017, then Hyundai acquired 80% stake in 2021.
-
2015: Introduced Spot series; commercialized in 2019; held 12.67% global market share by 2023.
-
2017: Launched Spot Mini combining wheeled and legged mobility.
-
2013–2024: Hydraulic-powered Atlas (1.5m tall, 28 DOF).
-
2024: Unveiled fully electric Atlas with added neck and waist rotation joints, achieving autonomous material handling in factory trials.
-
2025 Plan: Pilot production at Hyundai Motor factories in South Korea.
Technical Advantages
-
Drive System: Transitioned from hydraulic to full-electric drive, resulting in more compact structure and stable control.
-
Motion Control: High-speed movement and precise action in complex environments, with wide range of motion in waist and neck.
-
Hand Capability: Equipped with three-finger dexterous hand.
-
Environmental Perception: 360° sensing system adapted to complex industrial environments.
-
AI Collaboration: Jointly developed large behavioral models with Toyota Research Institute for complex decision-making tasks.
Main Achievements
-
Industrial Testing: Transported car parts in Hyundai factories in 2024.
-
Global Leader in Motion Control: Atlas regarded as the benchmark in humanoid robot motion control.
-
Clear Market Positioning: Focuses on high-value sectors like industrial inspection and hazardous operations.
-
Pilot Program: Launching mass production pilot in 2025, gradually expanding commercial use.
4. Current Challenges and Opportunities in Humanoid Robotics
Despite promising prospects, the humanoid robotics field faces dual technical and market challenges. Key bottlenecks include high hardware costs, insufficient AI understanding of the real world, and precision issues in motion control. Nevertheless, with continuous technological breakthroughs and accelerated capital investment, the industry holds immense growth potential.
-
Hardware and Cost Barriers: Tesla aims to bring Optimus down to $20,000 per unit, but current lab-grade robots still exceed $100,000, slowing commercialization.
-
AI and Sensor Technologies: Despite rapid AI advances, robots still lack sufficient generalization ability in complex scenarios, especially regarding real-time performance and accuracy in perception and decision-making.
-
Decentralized Networks: To enable broader adaptability, decentralized data networks require stronger real-time data processing capabilities.
Innovative solutions based on Web3 are rapidly emerging. Decentralized AI training and robotic economies can drastically reduce development costs while further promoting hardware and technology adoption. Just as Web3 demonstrated potential in the metaverse and AI domains, humanoid robotics may become the next core narrative for Web3.
II. Overview of Web3 Robotics and Embodied Intelligence Projects
1. Overview
With rapid advancements in AI, robotics, and blockchain, Web3 robotics projects are becoming pivotal intersections of emerging technologies. Like every transformative era awaiting the final puzzle piece, the technological tree for humanoid robots is now ready to be unlocked. By integrating embodied intelligence, decentralized physical infrastructure networks, token incentives, and AI agents, these projects aim to create an open, collaborative, self-driving “decentralized machine economy.”
Common characteristics of Web3 robotics and embodied intelligence projects include:
-
Real-World Task Execution: Robots collect environmental data via perception modules (e.g., cameras, LiDAR), guided by AI decision modules to perform tasks like navigation, monitoring, and data collection.
-
Decentralized Collaboration Mechanism: Token systems incentivize robot hardware providers, data contributors, and compute nodes, enabling cross-regional and cross-entity resource sharing.
-
AI Agent-Robot Integration: Treating AI agents as the “decision center” (brain) and robots as the “embodied executor” (body), achieving unified cognition and action.
-
Ecosystem Subdomains: Current Web3 robotics landscape can be divided into eight directions:
-
Robot Hardware Platforms
-
Robot Software & Protocol Layers
-
Decentralized Localization & Data Networks
-
Spatial Intelligence & Environmental Modeling
-
Infrastructure & Base Layer Networks (Layer-1)
-
Machine Economy Systems
-
DePAI DAO Organizational Forms
-
Gamified / Community-Based Robot Projects
-
2. Current Web3 Robotics and Embodied Intelligence Projects
(1) BitRobot
Project Overview
BitRobot Network is a decentralized network built on subnet (Subnet) architecture, aiming to accelerate embodied artificial intelligence (Embodied AI) through cryptographic economic incentives. Its core objective is to solve data and resource bottlenecks in the current Embodied AI domain by aggregating computing resources, robot swarms, datasets (real or synthetic), and AI models, thus driving scalable innovation in robotics.
Funding Status
In early 2025, FrodoBots Lab completed a $6M seed round in partnership with Protocol Labs to develop BitRobot. Led by Protocol VC, with participation from Big Brain Holdings, Fabric Ventures, Solana Ventures, and others. Co-founders of Solana and founders of several DePIN projects also invested.
Product Logic
Simply put, BitRobot is a blockchain- and token-incentivized “robot collaboration network,” allowing everyone to share robots, computing power, and data to jointly accelerate AI robot R&D.
Subnet Roles
-
Subnet Owners: Define subnet tasks (VRW), allocate reward ratios.
-
Subnet Validators: Verify validity of contributor outputs.
-
Subnet Contributors: Provide resources such as robot hardware, computing power, labor, etc.
Economic Model & Incentives
-
Network Rewards: Token incentives for resource providers like robot owners and compute suppliers.
-
Resource Utilization: Third parties (e.g., AI labs) pay to access subnet resources like robot clusters for model testing.
-
Commercialization: Open datasets and models available for non-commercial use; commercial licensing revenue reinvested into ecosystem.
(2) Reborn
Project Overview
Reborn transforms human motion data into tokenized assets used to train foundational robot models (RFM), thereby developing general-purpose humanoid robots.
Ecosystem Product Matrix
-
Unified Data Platform: Distributed infrastructure collecting high-quality real-world and synthetic data via motion-capture wearables, VR interaction, mobile video, and Roboverse simulation engine, laying foundation for scalable training.
-
Open Model Ecosystem: Reusable embodied AI model library including OpenVLA (vision-language-action model), full-body controllers, and dexterous manipulation models. Developers can directly deploy or extend via Reborn Physical AI App Store.
-
Physical Agents: Real-world robot agents planned to evolve through three stages: collaborative teleoperation → specialized model deployment → general autonomy, forming a data-driven practical pathway.
Network Roles
-
Data Contributors: Community users contribute VR/AR game data and real-body motion data to train RFM models and earn rewards.
-
Network Validators: Ensure data authenticity and usability, earning rewards.
-
Data Demanders: Robot developers purchase model usage rights or data.
Hardware Devices
According to the official website, the project has sold over 8,000 Rebocap™ devices, used for posture data capture and collection.
Data Categories
-
Embodied Vlog (real-world task videos): Captured using common cameras (smartphones, GoPro), showing first-person views of fine motor tasks like making sandwiches or washing dishes.
-
Mocap Life (precise motion capture via Rebocap™): Collects accurate human joint data.
-
VR Games: Partnerships with VR platforms to capture hand gesture and interaction data.
Funding Status
According to the website, the project received support from GGV Capital; specific funding details not disclosed.
(3) Vana
Project Overview
Vana is a decentralized protocol aimed at realizing data sovereignty and user-owned AI data layers. It allows users to own, control, and monetize personal data while building a DataDAO ecosystem. Vana treats data as a new class of digital asset, enabling users to pool data into DataDAOs for training AI models, breaking big tech’s data monopolies and promoting an open AI economy. Vana has attracted over a million users forming DataDAOs across health, robotics, science, prediction markets, and more.
Funding Status
Vana has raised $25M total, including a seed round led by Polychain Capital and Series A led by Paradigm, with participation from Coinbase Ventures. Funds are allocated to accelerate user-owned AI development, ecosystem expansion, and incubation of DataDAO projects.
Business Logic
Vana’s product logic centers on transforming data from a “extracted resource” to a “user-controlled asset.” Users upload encrypted data to Vana wallets, pooling it into DataDAOs to form Data Liquidity Pools (DLP). AI developers access these pooled datasets via smart contracts for model training, while data contributors receive governance tokens and rewards. This framework emphasizes privacy protection, collective bargaining, and market-driven value discovery, avoiding public validation pitfalls of traditional blockchains.
Technical Advantages
-
Privacy Protection: Encrypted data ownership records, access control, and verification proofs, supporting ZK tech for privacy assurance.
-
Scalable Network Architecture: Maintains global state for data ownership, access rights, and token balances; natively supports DataDAO pooling; uses VRC-20 standard for governance.
-
Interoperability: ERC20-wrapped version compatible with Ethereum ecosystem; integrates Ser and other infra services to optimize data access pipelines.
Robotics Narrative
Vana positions itself as critical data infrastructure in robotics networks, providing user-owned real-world video and sensor data to train physical AI and autonomous systems. It emphasizes that permissioned user data can scalably support robotics advancement, avoiding reliance on platform monopolies. Through DataDAOs, Vana aggregates diverse human data to improve robot-world interaction. Currently, Vana supports incubation of robotics ecosystem projects, including top-tier initiatives like PrismaX.
Token Status
Project TGE occurred in December 2024; current circulating market cap is $134M, fully diluted $525M, listed on Binance, Upbit, and other major exchanges.
(4) PrismaX
Project Overview
PrismaX is a decentralized data marketplace platform fueling real-world robot development. Through community-driven data collection and incentive mechanisms, it bridges the gap between robots and mainstream adoption. The project develops multimodal generative AI to help robots “see, understand, and interact” with the physical world via vision, video, and sensor data—not just text—aiming to overcome data bottlenecks and achieve autonomous, efficient coordination.
Business Model: Data Collection → Teleoperation → Model Training
-
B2B Data Services: Provides high-quality, diverse visual datasets to robotics/AI companies.
-
Teleoperation Platform: Offers standardized remote control solutions including operator management, payment, software interfaces, charged per usage or subscription.
-
Model Training & API Services: Partners with AI firms to offer pre-trained models or data augmentation services, earning licensing fees or revenue shares.
-
Decentralized Data Economy: Crypto incentives for data contributors, taking fees from data transactions.
Roadmap
-
Phase I: Teleoperation + Data Collection (2025–2026): Core goal—establish a globally distributed teleoperator network, primarily using robots for AI training data collection.
-
Phase II: Real Tasks + Edge Models (2026–2027): Core goal—robots perform real commercial tasks (logistics, manufacturing); operators manage multiple bots, AI-assisted to reduce latency.
-
Phase III: Full Autonomy + Machine Economy (2027+): Core goal—highly automated robot service network driven by foundational models.
Funding Status
Completed $11M raise in June 2025, led by a16z CSX, with participation from Virtuals Protocol, Volt Capital, Symbolic Capital, Stanford Blockchain Accelerator, among others.
(5) OpenMind
Project Overview
OpenMind’s OM1 is a modular AI runtime environment designed for agents and robots, featuring motion, voice, and other functionalities.
Main Features
OM1 allows AI agents to be configured and deployed in both digital and physical worlds. Create one AI agent—it can run in the cloud or execute on physical robot hardware, including quadrupeds, TurtleBot 3, and upcoming humanoid robots.
Use Cases
AI agents built on OM1 can integrate multi-source data (web, X/Twitter, cameras, LiDAR), enabling them to post tweets, explore rooms, shake hands, or converse. With OM1, you can not only chat with OpenAI’s gpt-4o but also literally “shake hands” with it.
Technical Features
-
Multi-Agent Endpoints: Allows different types of AI agents to collaborate via a single API.
-
Planner: Responsible for decision-making and task decomposition, setting goals and priorities.
-
Navigator: Plans execution paths, optimizes task flow.
-
Perception/Control: Handles environmental perception and physical operation execution.
-
-
Modularity & Scalability: Modular design allows developers to add or replace agents based on needs. This flexibility suits various applications—from robot control to complex decision systems.
Open Source Support
OpenMind provides partial open-source code on GitHub demonstrating recursive prompting for solution generation and integration with OpenAI’s GPT-4o API.
(6) FrodoBots
Project Overview
An innovative project focused on AI and robotics, dedicated to building an AI-powered humanoid robot fighting club and global robot network. Using DePIN technology, it creates an ecosystem combining robot competitions, gaming, and research.
Funding Status
Reportedly secured $2M pre-seed funding early on; in early 2025, FrodoBots Lab closed a $6M seed round with Protocol Labs to co-develop BitRobot.
Core Features
-
Robot Fighting Club: Develops humanoid robots for real-time combat, merging AI and entertainment.
-
Global Robot Network: Establishes a distributed robot fleet for data collection, urban navigation research, etc.
-
AI & Dataset Innovation: Generates unique datasets through robot battles to advance AI algorithms and robotics.
-
Community & Collaboration: Hosts AI workshops and demo events with global partners to promote tech exchange and adoption.
Economic Model & Token
Launched meme robot token $SAM @SamIsMoving earlier, a robot that collects real-world data in real time and periodically posts street view updates on its official X account.
(7) XMAQUINA
Project Overview
XMAQUINA is a DePAI DAO offering decentralized investment opportunities in robotics and machine finance. Holding platform token $DEUS grants access to investment opportunities and robot-generated income.
Main Vision
-
Build a Decentralized Machine Economy: Create a community-driven ecosystem enabling decentralized management and collaboration of machines and robots through partnership with peaq network.
-
On-Chain Robot Management Platform: Provides registration, authentication, and data management for robot devices, ensuring trusted operations in decentralized networks.
-
Community-Driven Governance & Distribution: Empowers community via $DEUS token and DAO mechanism to collectively determine rules and resource allocation in the machine economy.
Technology & Applications
-
Marketplace: Offers a decentralized marketplace for trading robot components, AI models, and services, fostering innovation and tech sharing.
-
Deus Labs: Internal R&D division of the DAO, focusing on open-source robotics, ethical AI, and decentralized tech development.
-
AI Agents: Develops AI agents to automate DAO operations and governance processes.
Funding Status
In late April 2025, IDO sold 25 million $DEUS tokens (2.5% of total supply of 1 billion $DEUS) at $0.04 per token.
In February 2025, community fundraising via DEUS Genesis Auction successfully raised 3,703,703 $PEAQ tokens on the peaq network—the first Genesis Auction on peaq.
Reportedly completed a seed round early on, with participation from EoT Ventures and Moonrock Capital; exact amount undisclosed.
(8) GEODNET
Project Overview
A decentralized real-time dynamic positioning network providing centimeter-level GPS accuracy for robots, drones, and automated machines. Builds AI-driven navigation systems and serves as a real-world geospatial information hub.
Working Principle
-
Base Stations: Individuals or enterprises purchase and install GNSS hardware (satellite miners) in sky-visible locations to collect satellite signals and upload data via internet. These stations calculate signal time differences to generate high-precision RTK correction data.
-
Blockchain & Token Incentives: Uses $GEOD token to reward participants. Base station operators earn $GEOD for high-quality GNSS data; data users buy RTK correction data access, reducing token circulation to support value.
-
Decentralized Verification: Employs Proof of Accuracy and Proof of Stake protocols to ensure data integrity and network security, preventing tampering.
Funding Status
February 2025: Closed $8M OTC round, led by Multicoin Capital, with ParaFi Capital, Digital Asset Capital Management, and others participating.
April 2024: Raised $2M, with Pantera Capital, CoinFund, VanEck, Santiago Roel Santos, and others participating.
Token Status
TGE in Q3 2023; current circulating market cap $57M, fully diluted $179M, listed on Gate, MEXC, and other exchanges.
(9) Auki/Posemesh
Project Overview
A DePIN-based spatial computing project aiming to build a decentralized machine perception network. Using collaborative spatial computing protocols, devices securely and privately share spatial data and computational power to form a shared understanding of the physical world.
Business Logic
-
Data Input: Any sensor-equipped device can join Posemesh to contribute spatial data; nodes provide compute and storage.
-
Data Processing: Posemesh integrates data to generate high-precision spatial maps and services.
-
Data Output: Enterprises/developers use data to optimize operations (e.g., robot navigation, AR experiences), paying in tokens.
-
Ecosystem Incentives: Data contributors and node operators receive token rewards; Auki Labs earns from transaction fees and custom services.
-
Positive Cycle: Token economy incentivizes more devices and nodes to join, expanding network scale and improving data accuracy and application scope.
Funding Status
Community whitelist presale around August 2024; according to official blog, 390 qualified addresses participated, raising ~$204,700 USDT.
Token Status
TGE at end of August 2024; current circulating market cap $69M, fully diluted $284M, listed on MEXC.
(10) peaq
Project Overview
peaq is a Layer 1 blockchain focused on DePIN and machine economy, founded in 2017.
Project Features
-
DePIN Infrastructure: peaq offers modular DePIN functions enabling developers to build decentralized physical infrastructure projects across IoT, mobility, energy, agriculture, gaming, and more.
-
Machine Economy: Assigns autonomous identity, payment capability, and access control to machines, enabling independent transactions, data exchange, and collaboration.
-
Cross-Chain Interoperability: Connected to over 90 blockchains, supporting cross-chain data and liquidity flows.
-
$PEAQ Token: Native network token used for transaction fees, governance, machine rewards, NFT ownership, and DeFi liquidity provision.
Main Functions
-
Decentralized Identity & Control (peaq ID and peaq Control): Assigns a self-managed digital identity (peaq ID) to each device for identification and authentication. peaq Control is a Web3 machine control center connecting devices, sensors, vehicles, and robots, ensuring secure network interaction.
-
Access Control & Permissions (peaq Access): Role-based access control lets developers define user groups, roles, and permissions to regulate access to devices or services. For example, only authorized users can access shared vehicles in mobility apps.
-
Decentralized Payments & Economics (peaq pay): Enables peer-to-peer payments between machines and between machines and users, ensuring fund security and transaction transparency. peaq pay uses multi-sig wallets to verify fund sufficiency and handles refunds or settlements after service completion.
-
Data Validation & Trustworthiness (peaq verify): Ensures data in DePIN networks is authentic and sourced from real physical devices, preventing fraud and tampering. Three-layer verification: machine origin authentication, data consistency checks, and community validation.
-
Modular DePIN Features & Developer Support: Offers a one-stop tech stack including smart contract support, cross-chain bridges, data storage, and AI agents.
-
Ecosystem Incentives (DePIN Flywheel): Designs incentive mechanisms via the “DePIN Flywheel” model—participants (device owners, developers, users) earn token rewards for contributing resources or data, attracting more devices and creating a virtuous cycle.
Funding Status
May 2024: Raised $20M via CoinList, with participation from over 14,500 community members.
March 2024: Closed $15M round led by Generative Ventures and Borderless Capital, with Spartan Group, HV Capital, CMCC Global, Animoca Brands, and others participating.
Token Status
TGE in November 2024; current circulating market cap $135M, fully diluted $483M, listed on Gate, Bitget, MEXC, and others.
(11) CodecFlow
Project Overview
CodecFlow is an AI operator and robot execution layer built on the Solana blockchain, focused on overcoming limitations of traditional automation in software and robotics. Using Vision-Language Actions (VLA) models, AI perceives, reasons, and executes actions on screens or robotic systems. The project includes Fabric (multi-cloud execution layer for GPU-intensive workloads) and Operator Kit (optr) (lightweight toolkit for building desktop, browser, simulation, and robot agents).
Product Logic
CodecFlow’s product logic revolves around end-to-end execution of AI operators: from perception (visual-language input) → reasoning (VLA model processing) → action (cross-screen/robot execution). Unlike traditional automation reliant on fragile UI scraping or APIs, CodecFlow uses Fabric to dynamically optimize compute resource allocation, selecting lowest-cost nodes for GPU tasks. Modular management of agent states, actions, and security ensures composability and safety. This logic emphasizes robustness in distributed computing, enabling seamless transition from local to remote execution, suitable for both software automation and physical robot interaction.
Core Features
-
VLA Model Integration: AI agents perceive environments, make decisions, and execute tasks via visual-language-action frameworks, supporting screen interaction and robot control.
-
Fabric Execution Layer: Multi-cloud GPU scheduling, real-time capacity and pricing sampling, selects optimal nodes for compute-intensive workloads.
-
Optr Toolkit: Lightweight SDK for building agents, providing action management, state tracking, simulation, and security modules, compatible with desktop/browser/robot environments.
-
Remote Simulation Support: Developers can run simulations on servers without high-end local hardware, enhancing accessibility.
Token Status
TGE in May 2025; current circulating market cap $30M, fully diluted $40M. Not yet listed on centralized exchanges, primarily traded on DEXs.
CA:69LjZUUzxj3Cb3Fxeo1X4QpYEQTboApkhXTysPpbpump
(12) RICE AI
Project Overview
RICE AI focuses on AGI robot development. Built on BNB Chain, it constructs a DePIN network connecting global robots for data sharing and collaboration, addressing data scarcity in robot AI development. Supported by Hong Kong-based Rice Robotics, the deployed Minibot M1 robot serves as its core hardware foundation. The platform aims to crowdsource high-quality real-world robot training data, democratizing the robotics industry and accelerating AGI development.
Funding Status
In August 2025, conducted $RICE token presale via TokenFi Launchpad, raising $750K (10% of total supply), with a project FDV of $7.5M.
Product Logic
RICE AI builds its product logic on the DePIN model: Global robots (including Minibot M1) upload sensor and interaction data to earn $RICE tokens. This data feeds into a decentralized AI foundry for training foundational robot models. Developers subscribe to models for AI app development, receiving discounts when paying with $RICE. The platform monetizes data via tokenization, solving high costs, delays, and silos in robot data collection, pushing the shift from simulated to real-world data, ultimately enabling autonomous robot learning and collaboration.
Core Features
-
Data Contribution & Rewards: Robot users upload interaction data (e.g., object grasping, motion trajectories) via VR teleoperation or automatic capture, earning $RICE.
-
Model Training & Access: Decentralized foundry trains physical AI foundation models using collected data; subscription access available with $RICE payment discounts.
-
Governance & Data Marketplace: $RICE used for platform governance and data trading; portion of platform fees used for token buyback and burn.
-
Robot Integration: Supports hardware like Minibot M1, with integrated GPU processing for real-time learning and collaboration.
-
Multi-Source Data Support: Includes third-party robots, teleoperation, egocentric videos, etc.
Token Status
TGE in August 2025; current circulating market cap $22M, fully diluted $120M. Listed on Binance Alpha, exchange listing on MEXC.
III. Conclusion and Outlook
Currently, the Web3 humanoid robotics and embodied intelligence field remains in early stages, with the overall ecosystem still forming and industry standards and application paradigms continuously evolving. Based on observed Web3 humanoid robotics project forms, they can be broadly categorized into three types:
First are new projects focused exclusively on Web3 humanoid robotics from inception—free of legacy constraints, able to natively integrate on-chain incentive mechanisms with embodied intelligence, offering strong imaginative potential.
Second are projects originally focused on DePIN, AI, or data domains that recently expanded narratives into robotics. These often have more mature teams and existing community traction, providing a solid foundation for rapid validation and deployment.
Third are lightweight projects centered on memes, gaming, or social elements—though technically less demanding, they possess strong community engagement and narrative flexibility, potentially producing breakout hits first.
In summary, despite facing multifaceted challenges in technology integration, user experience, and education, the paradigm shift brought by combining embodied intelligence with Web3 holds immense potential. As the most tangible embodiment of embodied intelligence, humanoid robots—once deeply integrated with on-chain identities, incentive mechanisms, and decentralized autonomous systems—could give rise to truly meaningful on-chain agents. These agents won’t just participate in the digital economy but could also reshape real-world production and collaboration. We firmly believe Web3 humanoid robotics will become a crucial narrative worth long-term attention, potentially serving as the key gateway connecting the physical and virtual realms.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
@dubailfg
