DePIN x AI: An Overview of Four Major Decentralized Computing Networks
TechFlow Selected TechFlow Selected
DePIN x AI: An Overview of Four Major Decentralized Computing Networks
This article explores the three largest general-purpose decentralized computing networks and one decentralized AI project, aiming to help readers understand the similarities and differences among these projects.
Navigating Web3 tides with focused insightsAuthor: 0xEdwardyw
-
Akash, Render Network, and io.net are the three largest decentralized computing networks in the market. Although they all offer decentralized computing services, each network has a different business focus.
-
Bittensor is a decentralized artificial intelligence project that leverages distributed computing resources for machine learning. Its goal is to directly compete with centralized AI services like OpenAI.
-
On the supply side, Akash has a diversified hardware network including CPUs, GPUs, and storage, while Render possesses a large number of GPUs. io.net aggregates substantial GPU capacity from its own network as well as other platforms.
-
Decentralized computing networks are two-sided markets, with each project’s token serving as a medium of exchange within their respective ecosystems. Render Network and Bittensor have implemented token burning mechanisms to enhance value accrual.
Types of Decentralized Computing Networks
How Akash Differs from Render Network
Both Akash and Render Network are decentralized computing networks that provide platforms where users can buy and sell computing resources for various tasks.
Akash operates as an open marketplace, allowing users access to CPU, GPU, and storage resources. It offers computing resources for diverse purposes such as hosting game servers or running blockchain nodes. On the Akash marketplace, tenants deploying applications set the price and conditions for deployment, and providers bid for these deployments. The lowest bidder (provider) wins the deployment. This reverse auction model empowers users to define pricing and terms.
In contrast, Render uses a dynamic pricing algorithm that adjusts task deployment prices based on market conditions. Render Network focuses specifically on GPU-based 3D rendering services and operates as a distributed GPU network. In this model, hardware providers contribute computing resources, and the Render network employs a multi-tiered pricing algorithm to determine prices and match users with service providers. Render does not operate as an open marketplace where users can independently set prices or conditions.
Io.net – Focused on Artificial Intelligence and Machine Learning
io.net is a new decentralized computing network that aggregates GPU computing power from distributed data centers, cryptocurrency miners, and decentralized storage providers to support machine learning and AI workloads. It also collaborates with existing decentralized computing networks like Render, leveraging underutilized GPU resources on Render to handle AI and machine learning tasks.
io.net's key differentiators are twofold: 1) a focus on AI and machine learning tasks; and 2) an emphasis on GPU clusters. GPU clusters refer to multiple GPUs working together as a unified system to handle computationally intensive tasks such as AI training and scientific simulations.
Bittensor – A Blockchain Project with an AI Focus
Unlike other decentralized computing networks, Bittensor is a decentralized AI project aiming to create a decentralized machine learning marketplace, enabling decentralized AI applications to be built and compete directly with centralized AI projects like OpenAI’s ChatGPT. The network consists of nodes (miners) that provide computing resources for training and running AI models.
Bittensor utilizes a subnet structure, where subnets resemble application-specific chains. Currently, it hosts 32 subnets, each focusing on specific AI-related tasks, including decentralized text-to-text AI networks (similar to ChatGPT), text-to-image AI generators, and AI-powered search engines.
Miners play a crucial role in the Bittensor ecosystem by providing computing resources and hosting machine learning models to perform off-chain AI computations and generate results. Anyone meeting the minimum hardware requirements can join the network as a miner. Miners compete with one another to deliver the best results for user queries.
Network Capacity and Usage
Akash initially focused on CPUs and has accumulated a significant amount of CPU resources within its network. With the rise of AI, demand for GPUs has surged dramatically. Since last year’s third quarter, Akash has been adding GPU resources to its network. However, compared to projects focused specifically on GPUs, Akash has relatively fewer high-performance GPUs. Render Network specializes in decentralized, GPU-based rendering solutions, enabling it to amass a large number of GPUs across its network.
Render Network and Akash are more mature projects, with network usage growing year over year. Particularly after expanding its focus to include GPUs, Akash has seen a significant increase in quarterly active leases.
io.net is a newer decentralized computing network that launched its public testnet in November 2023. Despite its short history, io.net has aggregated substantial GPU capacity by integrating resources from Render, Filecoin, and its own network. Recently, io.net announced support for Apple Silicon chip clusters, enabling Apple users to contribute their unused computing power to the network, further increasing its available hardware. Additionally, io.net has not yet launched its protocol token, and many hardware providers may be joining the network in anticipation of potential token airdrops.
Bittensor is a decentralized AI network where miners contribute computing resources. Miners can either invest in their own hardware setups or simply use cloud-provided computing resources. In terms of hardware scale, Bittensor cannot be directly compared to typical decentralized computing networks. Currently, Bittensor has over 7,000 miners.
Tokenomics
Decentralized computing platforms function as two-sided markets where users pay computing resource providers. Akash, Render Network, and Bittensor have all issued their own tokens as mediums of exchange within their ecosystems. Render and Bittensor have implemented token burn mechanisms to enhance value accumulation.
Akash
Akash is an independent PoS blockchain, with $AKT as its native token used for staking to secure the network and pay transaction fees. The token also serves as a medium of exchange within the ecosystem, with $AKT being the primary pricing unit for transactions and leases on Akash. As a PoS chain, Akash requires issuing $AKT to provide block rewards for validators, with a current inflation rate of approximately 14%.
Currently, Akash charges a 4% fee on transactions paid in AKT and 20% for those paid in USDC. These fees go into a community treasury. The specific use of treasury funds has not yet been determined and could include public funding initiatives, incentive programs, or potentially token burns.
Render Network
Render Network has migrated from Ethereum to Solana, and its protocol token RNDR is used for value exchange within the Render ecosystem. Creators and users pay for rendering jobs using this token.
To balance the dynamics between supply and demand for computing resources, Render implements a Burn-and-Mint Equilibrium (BME) mechanism. When demand (i.e., rendering jobs) exceeds supply, RNDR tokens are burned, creating a deflationary effect. Conversely, if supply exceeds demand, additional RNDR tokens are minted, leading to inflation. Currently, due to insufficient computational demand, RNDR is experiencing inflation.
Bittensor
Bittensor’s native token $TAO is used to access network services and as the core medium for reward distribution. $TAO has a maximum supply of 21 million, with 7,200 tokens generated daily as rewards for miners and validators. Bittensor implements a halving mechanism: once half of the total supply has been distributed, the issuance rate will halve. After the first halving, subsequent halvings occur when half of the remaining supply is distributed, continuing until the 21 million cap is reached.
Although the current issuance rate of 7,200 TAO per day is fixed, the timing of the next halving is not predetermined due to the token recycling mechanism. This mechanism burns issued TAO tokens, effectively delaying the point at which half of the total supply is distributed. Miners and validators must recycle (i.e., burn) TAO tokens to register on the network. These burned tokens are removed from circulation but can be re-mined later. The network periodically deregisters miners and validators that fail to deliver competitive AI results, requiring them to pay/burn TAO again upon rejoining, making registration a recurring cost. This dynamic burn mechanism creates sustained demand for TAO.
The originally planned first halving was scheduled for January 2025, but the current halving date has been delayed to October 2025, indicating that a significant number of TAO tokens have already been burned.
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
@TokenInsight
