What is IO?
Introduction to IO
IO Coin (abbreviated as IO) is the native currency of the IOG network. Within the IOG network, IO is used to balance the economic incentive needs of three main groups within the ecosystem:
The first group is GPU renters (i.e., users), including machine learning engineers who want to purchase GPU computing power on the IOG network. These engineers can use IO to deploy GPU clusters, cloud gaming instances, and build pixel streaming applications similar to Unreal Engine 5. Additionally, users include individual consumers who wish to conduct serverless model inference on BC8.ai and hundreds of applications and models that will be hosted on io.net in the future.
The second group is GPU owners (i.e., suppliers), such as independent data centers, crypto mining farms, and professional miners who wish to supply and monetize their underutilized GPU computing power on the IOG network.
The third group is IO Coin holders (i.e., the community), who participate by providing cryptoeconomic security and incentive mechanisms to ensure mutual benefits and penalties among all parties, thereby promoting network growth and adoption.
It is worth noting that these groups are not mutually exclusive; an IO holder can also be a GPU renter or a GPU owner, or both.
IO Tokenomics
IO tokenomics refers to the overall structure and mechanism of the IO token within its ecosystem, involving token distribution, supply and demand dynamics, utility within the ecosystem, stakeholder incentives, governance mechanisms, and mechanisms for value capture and distribution. Here is a detailed introduction to the tokenomics of io.net.
Firstly, IO has a fixed maximum supply of 800 million tokens.
At the initial launch, 500 million IO tokens will be distributed, with the remaining 300 million tokens to be released as rewards to suppliers and their stakers hourly over 20 years. Rewards follow a disinflationary model, starting at 8% in the first year and decreasing by 1.02% per month (approximately 12% per year) until the 800 million IO cap is reached.
To ensure the scarcity of IO tokens, IO also adopts a burn mechanism. Revenue generated by io.net from the IOG network is used to purchase and burn IO tokens. The burn mechanism adjusts the number of IO tokens to be burned based on the token's price, thereby reducing the circulating supply and creating deflationary pressure.
io.net generates revenue by charging fees to users and suppliers. Users need to pay a 0.25% reservation fee when booking computing power. For payment fees, a 2% fee is charged for 100% USDC payments, while there are no fees for 100% IO payments. Suppliers also pay a 0.25% fee when their nodes are booked, deducted from their computing power earnings. Similarly, suppliers using 100% USDC payments pay a 2% fee, while those using 100% IO payments incur no fees.
What is IO Network
IO Network (io.net) is the world’s largest decentralized AI computing network, allowing machine learning engineers to access scalable distributed clusters at a fraction of the cost of traditional centralized services. Unlike conventional centralized services, io.net can create tens of thousands of GPU clusters with lower latency, whether they are co-located or geographically distributed.
In the IOG network, besides direct suppliers, decentralized physical infrastructure networks (DePINs) like Render (focused on image rendering) and Filecoin (focused on storage) also supply their computing power to io.net to monetize from AI/ML companies.
io.net establishes an ecosystem of products and services by using compute as currency, reducing costs for AI/ML innovators and promoting the adoption of AI technology. Our vision is to make IO the currency of compute, supporting a range of products and services so that compute resources can be used as both a resource and an asset.
Modern machine learning models frequently leverage parallel and distributed computing, utilizing multiple cores across various systems to optimize performance or scale to larger datasets and models. However, the capacity of traditional cloud service providers falls far short of the market demand from AI/ML companies, leading to several challenges:
1. Limited Availability: It can take weeks to access hardware using cloud services like AWS, GCP, or Azure, and popular GPU models are often unavailable.
2. Limited Choices: Users have few options regarding GPU hardware, location, security level, latency, and other factors.
3. High Costs: Acquiring high-performance GPUs is extremely expensive, and projects can easily spend hundreds of thousands of dollars per month on training and inference.
io.net addresses these issues by aggregating underutilized resources from independent data centers, crypto miners, and other hardware networks. These resources are combined within a Decentralized Physical Infrastructure Network (DePIN), providing engineers with massive on-demand computing power that is accessible, customizable, cost-efficient, and easy to implement.
With io.net, teams can scale their workloads across a network of GPUs with minimal adjustments. The system handles orchestration, scheduling, fault tolerance, and scaling, supporting various tasks such as preprocessing, distributed training, hyperparameter tuning, reinforcement learning, and model serving. It is designed to serve general-purpose computation for Python workloads, particularly AI/ML workloads.
The four core functions provided by io.net are:
1. Batch Inference and Model Serving: By exporting the architecture and weights of trained models to shared object storage, io.net allows machine learning teams to build inference and model-serving workflows across a distributed network of GPUs.
2. Parallel Training: CPU/GPU memory limitations and sequential processing workflows present a significant bottleneck when training models on a single device. io.net leverages distributed computing libraries to coordinate and batch-train jobs, enabling parallelization across multiple distributed devices.
3. Parallel Hyperparameter Tuning: Hyperparameter tuning experiments are inherently parallel. io.net uses distributed computing libraries for advanced hyperparameter tuning, simplifying scheduling and search patterns, and checkpointing the best results.
4. Reinforcement Learning: io.net employs an open-source reinforcement learning library, supporting production-level, highly distributed RL workloads with a simple set of APIs.
Official Links of IO
Website: https://io.net/
Twitter: https://twitter.com/ionet
Discord: https://discord.com/invite/ionetofficial
Telegram: https://t.me/+vr2_-Vr9YOZhNDNh
Steps to Buy IO on LBank
Purchasing IO tokens on the cryptocurrency exchange platform LBank is straightforward. Here's a step-by-step guide:
1. Log into your account: Ensure you're registered and logged into your LBank account, then click the "Trade" option in the top menu bar to start the trading process.
2. Search for IO: Use the search function within the trading platform to find "IO", locate the trading pair, such as IO/USDT, and click to enter its dedicated page.
3. Place an order: In the IO/USDT trading page's spot trading area, enter the amount of IO you wish to buy. If you want to complete the purchase immediately at the current market price, select a market order and then click the "Buy" button.
4. Confirm the order: A confirmation window will pop up. After verifying the order details are correct, confirm to place your order. Once the transaction is successful, the purchased IO will automatically be deposited into your LBank spot wallet.