Introduction: The d/acc Movement and Decentralized AI

In “My techno-optimism” published last November, Ethereum founder Vitalik introduced the philosophy of d/acc (defensive, decentralized, democratic, differential acceleration). This approach emphasizes selectively advancing technology to avoid over-concentration and monopolies, ensuring technological progress benefits all of society.

In AI (Artificial Intelligence), Vitalik highlighted the potential for AI to surpass human intelligence and become a new dominant species, necessitating a cautious approach to its development. Current AI governance structures have issues, with OpenAI linked to Microsoft, Anthropic to AWS and Google, and a few tech giants monopolizing GPU power and user data.

“We need a truly open AI ecosystem.” This belief has driven many Web2 AI entrepreneurs to shift to Web3, seeking decentralized and democratic solutions.

The Founding of NetMind.AI

Kai Zou, founder and CEO of NetMind.AI, is an innovator dedicated to merging AI with Web3. With extensive experience in the Web2 field, he believes that as AI models grow in scale and application, distributed and decentralized AI is inevitable. He sees the Web3 framework as a crucial solution providing broader opportunities for AI development.

Kai Zou graduated from Tsinghua University in Mathematics and Physics in 2010 and earned a master’s degree in Mathematics and Statistics from Georgetown University in 2013. Inspired by DeepMind and OpenAI, he founded the AI company ProtagoLabs in 2016, serving as CEO. This company, more like an AI research studio, has had its research papers published at top conferences such as AAAI and IEEE. Among these, the paper “EDA: Easy Data Augmentation,” co-authored with OpenAI researcher Jason Wei, has been cited over 2,000 times.

Beyond academic achievements, Kai Zou has invested in several AI startups, including Haiper.ai, Auto Edge, Qdot, and Orbit. In 2021, he decided to create a product-focused company. While training and fine-tuning language models, he realized the high cost of GPUs made them unaffordable for ordinary AI enthusiasts and small research institutions. Thus, he began researching asynchronous training architectures to combine small GPUs for neural network training, leading to the creation of NetMind Power, a platform encouraging people to contribute to idle machines to form a large distributed computing network.

Kai Zou firmly believes that only through decentralization and open collaboration can AI technology truly benefit humanity.

NetMind Power & NetMind Chain

NetMind Power is the core product of NetMind.AI, a decentralized AI full-stack development platform based on NetMind Chain.

How NetMind Chain Works

NetMind Chain uses the Proof of Authority (PoA) consensus mechanism, with 21 master nodes responsible for verifying transactions and generating blocks. These master nodes are elected by the community and must be reputable and continually contribute to the community. Compared to mechanisms like Proof of Work (PoW) and Proof of Stake (PoS), PoA significantly reduces energy consumption and maintenance costs while ensuring decentralization and security.

The NetMind Chain ecosystem includes the following participants:

• Compute Providers: Individuals or institutions contributing idle GPU compute resources, earning NMT rewards based on effective work.
• Compute Users: Developers and researchers using the compute resources of the NetMind Power platform for AI model training, fine-tuning, and inference, paying NMT tokens as service fees.
• Stakers: Supporting master node operations by staking NMT tokens and earning staking rewards.
• Liquidity Providers: Providing liquidity for NMT on decentralized exchanges and earning transaction fees.

To efficiently aggregate globally distributed GPU resources, the NetMind Power platform uses innovative decentralized cluster technology. When a user initiates a model training, fine-tuning, or inference task on the platform, it is automatically assigned to the most suitable GPU resources with the lowest latency. NetMind.AI employs peer-to-peer networks, containerization, Kubernetes, and other technologies to achieve unified scheduling and management of large-scale heterogeneous compute resources.

Additionally, NetMind.AI is developing asynchronous training technology to tackle network bottlenecks in decentralized environments. Traditional distributed training requires real-time communication between GPUs via high-speed networks, which is challenging in decentralized scenarios. NetMind.AI’s asynchronous training splits training data and AI models into smaller parts, distributes them to different GPUs for parallel processing, and aggregates the results, reducing the demand for network bandwidth while maintaining training effectiveness.

Finally, to ensure privacy and security, NetMind Power adopts model encryption and data isolation technologies. The platform encrypts AI model code, training data, and model weights at each node and during transmission. Through data partitioning and model segmentation, no single node can access the complete dataset or model, greatly enhancing security and privacy protection.

NMT Token Economics Model

NetMind Token (NMT) is the native utility token of the NetMind Chain, designed for various platform functions, including paying for training and inference services, rewarding users for contributing computing resources, and participating in platform governance.

The total supply of NMT is 147,571,163 tokens. The platform controls the issuance and burning of tokens through smart contracts to regulate market supply and demand. For example, the platform adjusts the allocation ratio of computing fees based on market demand and the NMT price to ensure token value stability.

On April 16, 2024, NetMind.AI launched a new token economic model. Over the next 10 years, 113 million NMT tokens will be issued, with over 50% allocated to the platform’s computing power providers.

Token Allocation (10-Year Total):

• 62.5 million NMT (55.31% of new issuance, 42.35% of new total supply) for computing power rewards, incentivizing contributors to the network.
• 16.5 million NMT (14.60% of new issuance, 11.18% of new total supply) for staking rewards, incentivizing users to stake and provide liquidity.
• 16.5 million NMT (14.60% of new issuance, 11.18% of new total supply) for the ecosystem growth fund, to expand and enrich the NetMind.AI ecosystem.
• 7.5 million NMT (6.64% of new issuance, 5.08% of new total supply) reserved for strategic investors and advisors.
• 10 million NMT (8.85% of new issuance, 6.78% of new total supply) for the team fund, rewarding the NetMind.AI technical team.

As more users choose to hold and stake NMT long-term for earnings, the actual circulating supply of NMT will significantly decrease. Currently, over 20 million NMT are staked, accounting for 57% of the circulating supply.

Through a well-designed token economic model, NetMind.AI closely ties economic incentives to system contributions, creating a virtuous ecosystem: the more people use the NetMind Power platform, the greater the demand for NMT, stabilizing its price, and attracting more computing power providers and developers. NetMind.AI is exploring a new, sustainable path for decentralized AI development.

Stable Computing Power Supply and Demand

As Nvidia’s revenue and stock price continue to hit new highs, the notion that “computing power is the new oil of the AI era” has taken root. Decentralized computing power leasing has become a highly competitive field, with several similar projects emerging, such as io.net, Aethir, Akash, Gensyn, and Golem.

To evaluate the product-market fit (PMF) of these computing power leasing platforms, both supply and demand dimensions must be considered:

• Computing Power Supply: Can the platform aggregate enough GPU computing power resources?
• Computing Power Demand: Is there sufficient demand for the computing power aggregated by the platform?

Continuously Growing GPU Computing Power

According to Official dashboard data, the GPU computing power of the NetMind Power platform has significantly increased since February. Currently, 2,076 GPUs are connected to the platform, with another 4,804 GPUs queued to join.

The NetMind Power platform offers various GPU specifications and models, ranging from mid-to-low-end GPUs (such as RTX 4070 Ti, V100, RTX 4060 Ti, RTX 3070, RTX 3060) to high-end GPUs (such as 3090 Ti, 4090, A100, and H100).

Notably, high-end GPUs on the NetMind Power platform have very high utilization rates, especially the A100 and 4090, with usage rates nearing full capacity (98% and 99%, respectively). This indicates a high demand for high-performance GPUs on the platform, mainly for complex and computation-intensive tasks.

Diverse Partner Network

Partners use over 1,000 GPUs on the NetMind Power platform, accounting for more than 50% of the platform’s total computing power. This success is attributed not only to technological advantages but also to an extensive and deep partner network.

In the business sector, NetMind.AI has established strategic partnerships with several innovative AI companies, such as ProtagoLabs, Haiper.ai, Auto Edge, Qdot, and Orbit. These partners leverage the computing resources and AI ecosystem of the NetMind Power platform to accelerate their product development and iteration.

The NetMind.AI team has a strong AI background, with founder Kai Zou having published multiple papers at top academic conferences and collaborated with OpenAI researchers. This background enables NetMind.AI to connect with professional AI researchers and institutions. Currently, NetMind.AI has established academic partnerships with renowned universities, including Tsinghua University, Cambridge University, University of Wisconsin-Madison, Shanghai Jiao Tong University, Huazhong University of Science and Technology, and Fudan University. These collaborations involve cutting-edge AI research projects in areas such as graph neural networks, neural architecture search, natural language processing, and data privacy protection, fully utilizing the computing power and AI ecosystem provided by the NetMind Power platform.

In April this year, the NetMind Power platform began supporting fiat payments, further lowering the user threshold. Users can purchase computing power and services directly using credit cards or bank transfers without needing cryptocurrency exchanges. This move not only facilitates traditional enterprise customers but also attracts more individual developers and AI enthusiasts to the platform.

Model as a Service (MaaS)

NetMind.AI not only provides computing power leasing but also aims to offer full-stack AI development services, including an open-source model library, high-quality datasets, and an easy-to-use training and inference platform. This Model as a Service (MaaS) model makes AI development easier and more efficient. Developers can fine-tune and deploy pre-trained models directly on the NetMind Power platform, significantly reducing development costs and barriers.

Notably, datasets under NetMind.AI have been used as benchmarks for Google’s Gemini models. In the AGI Odyssey’s mathematics benchmark test, Google DeepMind’s latest AI model Gemini 1.5 Pro achieved a breakthrough score of 55.8%. This dataset, originally designed for the global AI competition (GAIC) hosted by AGI Odyssey and technically supported by NetMind.AI, is now open-source on GitHub:

https://github.com/protagolabs/odyssey-math/tree/main。

Additionally, NetMind.AI supports users in uploading their models to provide services to others, forming a virtuous ecosystem. This comprehensive AI ecosystem ensures that the aggregated computing power is fully utilized without concerns of excess capacity.

In summary, NetMind.AI has established a strong moat in both computing power supply and demand. On one hand, innovative technical solutions and token economic incentives have aggregated a significant amount of global idle computing power resources; on the other hand, NetMind.AI’s unique AI ecosystem and top-tier partner network have found a real demand for these computing resources.

Comprehensive AI Ecosystem

Beyond the decentralized AI full-stack development platform NetMind Power, NetMind.AI provides a range of decentralized AI products and services to build a complete AI ecosystem, making AI accessible and affordable for everyone. As founder Kai Zou stated, “We are working on all levels. I know it sounds ambitious, but we are working on all levels. All levels are ready and open to the public.”

NetMind Chat & Gaia

NetMind Chat is an AI chatbot platform where users can explore and choose the large language model that best suits their needs. The platform offers a simple chat interface, allowing users to compare different language models and fine-tune these open-source models through the NetMind Power platform.

In addition to well-known open-source language models, NetMind Chat features Gaia, an internally developed chatbot. Gaia is based on the Generative Pre-trained Transformer family and has been fine-tuned with supervised learning and reinforcement learning techniques on datasets collected by NetMind.AI. Gaia also uses a retrieval-enhanced method, enabling it to access extensive online knowledge beyond the model’s internal parameters.

Digital Human Avagi

Avagi is a customizable digital human model that lets users design realistic appearances, actions, and expressions. Users can animate the digital human using voice, text, or gestures for various purposes, such as entertainment, education, marketing, or social media. The platform includes embedded AGI (Artificial General Intelligence) digital entities, offering immersive conversational experiences through intelligently generated video clips. NetMind.AI is also developing a 3D version, which will be launched soon.

AI Life and Healthcare Solutions NetMind Life

NetMind Life leverages AI to provide personalized health solutions, helping users lead longer, healthier lives. Through AI-driven comprehensive assessments, it offers tailored lifestyle advice, health monitoring, nutritional supplements, and treatment plans.

AI Agent Development Tools and Platform NetMind XYZ

NetMind XYZ is an AI agent development tool and platform that provides users with all the modules needed to easily build AI agents. NetMind.AI will also develop multi-agent systems on the platform, enabling AI agents to communicate and collaborate on projects.

Inspired by founder Kai Zou’s vision of AI agents and multi-agent systems: “I believe that in the future, humans and AI agents will coexist in society. AI agents might have needs similar to humans; they might need to go to school to learn, work in companies to earn money, and communicate with each other to evolve into better intelligent agents.”

Currently, NetMind.AI is creating a fully AI-based marketing company, which will be the first multi-agent system on the platform.

AIGC Tools Suite NetMind Studio

NetMind.AI plans to offer a suite of AI-generated content (AIGC) tools, encompassing text, images, audio, and video, to help creators boost productivity and creativity.

By developing competitive products across computing power, models, data, tools, and applications, NetMind.AI is building a comprehensive and robust AI ecosystem. This full-stack approach not only better meets the diverse AI needs of users but also creates synergistic effects among different products and services, resulting in a stronger competitive edge.

Conclusion: The Boundless Potential of Decentralized AI

According to CoinGecko, the current market capitalization of AI tokens is only $35.4 billion. This is relatively small compared to other sectors. For instance, the market cap of DeFi is $105.6 billion, and the market cap of L1 public blockchains is a staggering $2.16 trillion. In the Web2 realm, the top 47 AI companies have a combined market cap of $9.945 trillion. These statistics show that the decentralized AI sector has vast development potential and growth opportunities. We are on the brink of an AI Summer.

NetMind.AI is poised for significant growth compared to other leading decentralized AI projects. This potential comes from two main factors: the deep technical expertise and professional background of the NetMind.AI team in AI, and the comprehensive AI ecosystem that NetMind.AI has built, which spans from computing power leasing to model training, inference, and deployment. This dual strength in technology and ecosystem positions NetMind.AI uniquely in the decentralized AI landscape.

As the NetMind.AI ecosystem continues to evolve and its user base expands, the demand for its native token NMT will increase, enhancing its value.

Disclaimer:

1. This article is reprinted from [theblockbeats], with all copyrights belonging to the original author [思维怪怪]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly according to the relevant procedures.
2. The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned otherwise, copying, distributing, or plagiarizing the translated articles is prohibited.