DePin

Both GEODNET and traditional CORS networks deliver RTK high-precision positioning services, yet GEODNET operates on a Decentralized Physical Infrastructure Network (DePIN) model, whereas traditional CORS networks are predominantly centrally built and maintained by government agencies, surveying authorities, or commercial operators. While both depend on GNSS reference stations to produce positioning correction data, they differ markedly in network expansion approaches, cost structures, and participation models.

GEODNET (GEOD) is a global high-precision positioning network built on a Decentralized Physical Infrastructure Network (DePIN). By incentivizing users to deploy GNSS reference stations, it delivers centimeter-level Real-Time Kinematic (RTK) services for drones, robots, autonomous driving systems, and surveying equipment. Unlike traditional centralized positioning services, GEODNET leverages blockchain and token incentives to expand global positioning infrastructure, enabling high-precision coverage of a wider area at a lower cost.

RTK (Real-Time Kinematic, 实时动态定位) is a high-precision positioning technology built on Global Navigation Satellite Systems (GNSS). By calculating satellite signal errors in real time at a reference station and transmitting correction data to user devices, it elevates ordinary GPS accuracy from meter-level to centimeter-level precision. GEODNET integrates RTK with a decentralized physical infrastructure network (DePIN), leveraging globally distributed GNSS reference stations to continuously generate and share correction data. This delivers high-precision positioning services with broader coverage and lower cost. Unlike traditional RTK networks, GEODNET employs token-based incentives to fund infrastructure growth, enabling community-driven global expansion.

GEODNET and Helium are both part of the DePIN (Decentralized Physical Infrastructure Network) sector, but the infrastructure they build serves different purposes. At their core, GEODNET provides location data infrastructure, solving how devices pinpoint their position with accuracy, while Helium offers connectivity infrastructure, addressing how devices connect to the network. Both leverage token incentives to advance real-world infrastructure, yet they differ significantly in their target users, data types, business models, and industry applications.

IO (io.net) is a decentralized GPU compute network built for artificial intelligence (AI) and machine learning (ML) applications. By pooling underutilized GPU resources worldwide, it delivers on-demand, high-performance computing power to developers, enterprises, and AI projects.

Energy asset tokenization refers to the use of blockchain technology to map real-world energy infrastructure, such as electric vehicles, battery energy storage systems, and solar equipment, into verifiable on-chain digital assets. Through device identity verification, data recording, and value settlement mechanisms, energy devices can not only participate in energy production and consumption, but also become value nodes within digital economy networks.

OpenVPP’s operating process mainly includes six stages: device registration, data collection, energy resource aggregation, demand response coordination, on-chain value calculation, and energy settlement. Through blockchain and smart contracts, OpenVPP digitalizes and opens up the energy management process used in traditional virtual power plants, providing infrastructure support for the energy internet and energy asset tokenization.

A Virtual Power Plant（VPP）is a management model that uses digital technology and intelligent control systems to aggregate distributed energy devices into a unified energy network. By connecting distributed energy resources such as solar power systems, energy storage devices, electric vehicles, and smart meters, a VPP can participate in power dispatch, grid balancing, and energy market trading much like a traditional power plant.

OpenVPP is a decentralized virtual power plant network built for distributed energy resources（DERs）. Through blockchain technology, it connects real-world energy infrastructure such as electric vehicles, energy storage systems, solar equipment, and smart meters. The network aims to create a unified energy data layer, device identity layer, and value settlement layer, allowing energy assets to be recorded, coordinated, and incentivized on-chain in a way similar to digital assets.

IO’s core use cases are concentrated in fields with high demand for GPU computing power, including AI model training, AI inference services, machine learning development, Web3 infrastructure, and DePIN network building. Compared with traditional cloud computing platforms, IO aims to aggregate idle GPU resources around the world and provide developers with a more flexible way to access computing power.

The development of AI infrastructure is driving continued growth in global demand for GPU computing power. As large language models, AI Agents, and inference services enter a phase of rapid expansion, GPUs have gradually become an important productive resource in the digital economy. Compared with traditional cloud service providers, which rely on self-built data centers to deliver computing power, IO aims to build a decentralized compute network by aggregating idle GPU resources around the world. The IO Token is the core economic asset that supports the operation of this network.

Janction is a decentralized computing network built for the age of artificial intelligence. By integrating distributed computing resources, AI Agents, and blockchain based incentives, it provides open infrastructure for AI model training, inference, and intelligent task execution. Janction aims to enable the discovery, allocation, coordination, and value settlement of computing resources without relying on centralized cloud service providers.

Caspius and traditional AI data platforms both serve AI model training, but they differ clearly in data ownership, value distribution, and network structure. Traditional AI data platforms usually follow a centralized model, with companies collecting and managing training data in a unified way. Caspius, by contrast, uses blockchain-based incentive mechanisms to build an open data contribution network, allowing users to participate in the collection and sharing of robotics training data.

Caspius provides AI models with the data they need for training by incentivizing users to upload first person videos, motion trajectories, and real world environment interaction data. Compared with traditional centralized data platforms, Caspius places greater emphasis on open data contribution and on chain incentive mechanisms. Robotics AI and Physical AI systems require large amounts of real world behavioral data to learn action execution, environmental understanding, and spatial interaction. Through a decentralized network, Caspius aims to expand the supply of robotics training data and provide more scalable data infrastructure for AI Agents, robotic systems, and automation devices.

Caspius is a decentralized AI data infrastructure protocol designed for embodied AI. It is mainly used to collect and distribute real world data needed for robotics training. By incentivizing users to upload first person videos, motion trajectories, and environment interaction data, Caspius aims to build an open robotics training data network that provides foundational data support for robotic models, automation systems, and Physical AI.