1. What is DIN

DIN is the first AI agent blockchain. Based on the Data Intelligence Network, it aims to provide comprehensive solutions and infrastructure for AI agents and decentralized AI applications (dAI applications).

1.1 Next-generation data and artificial intelligence blockchain network

As artificial intelligence technology continues to develop, we identify two key trends. First, AI agents will fundamentally change the way users interact with software, replacing multiple applications or decentralized applications (dApps) to form a unified personal assistant. This assistant communicates with users in natural language, interprets user intent, translates these intentions into executable code, and executes them using the best strategies. Second, to continuously improve the capabilities of AI agents, a large amount of data is needed for AI training. This data includes public and private user data. A large amount of cleaned public data is crucial for AI agents to make accurate decisions; similarly, the more personal data AI agents can access, the better they can accurately meet customized user needs (we do not delve into the potential privacy issues related to data authorization here).

1.2 DIN’s long-term vision

Building the infrastructure for AI agents and AI decentralized applications (AI-dApps). DIN’s value proposition is to build a data intelligence network based on the collaboration of humans, data, and AI. By using blockchain as the recording and settlement layer, we aim to establish a primary data layer for collecting, verifying, vectorizing, and incentivizing on-chain and off-chain data.

In the future, we believe that AI agents will fundamentally change the way users access information and interact with blockchain. AI agents with various capabilities will provide users with fully customized information and assist users in executing complex operations. The iterative improvement of AI capabilities requires everyone to contribute high-value data. With the continuous increase of data, AI agents will be able to access more and more information, making the network more intelligent. A more intelligent network will in turn help humans solve increasingly complex problems, benefiting data contributors and developers. The three core elements of data, people, and artificial intelligence will collaborate, promote each other, drive the positive iteration of the data intelligent network, and continuously enhance the value of the network. Let us together embrace this new era shaped by artificial intelligence and encryption technology, which is collectively built by everyone and benefits everyone. So, are you ready for the new era of artificial intelligence and data revolution?

2. Overview of current AI trends and markets

The rapid development of Artificial Intelligence (AI) technology has seamlessly integrated into our daily lives, fundamentally changing the way we interact with the world. DIN stands at the forefront of this revolution, aiming to build an intelligent network driven by data and AI agents. In the following sections, we will skip the general stages of AI and directly discuss Large Language Models (LLM) and AI agents. The reason is simple: whether in academia or industry, the development of AI will ultimately point towards one direction - AI agents. Humanity has entered a new era - the era of AI agents. In this era, AI transcends its role as a tool or service provider in the past, further developing into intelligent agents capable of active learning, adaptability, and autonomously undertaking complex tasks.

2.1 The gaps and opportunities in the market

In the data and its derivatives industry (including data indexing, storage, and analysis), it is widely believed in the industry that there is still huge untapped potential in the cryptocurrency field. With the popularity of various Layer 2 solutions and modular blockchain architectures, this potential is particularly evident, promising to enrich the application ecosystem built on these architectures. In addition, the advancement of AI technology will significantly increase the demand for data, especially structured data and AI agent demand. The use of data in AI scenarios is expected to become increasingly complex, mainly manifested in two aspects:

1. The increasing integration with traditional financial markets: The approval of BTC ETF signifies that relying solely on on-chain data is no longer sufficient for comprehensive data analysis. Traditional finance, stock, bond markets, and even national-level policies are increasingly closely connected with the cryptocurrency field. Information and data are multidimensional, requiring alignment and analysis based on timestamps to reveal potential signals hidden in different data sources.
2. The improvement of AI agents: The computing power of AI agents has reached the level where they can analyze highly complex data, and the parameters of large models have now reached the trillions. Now, data analysis tasks that surpass human capabilities have entered the realm of AI.

This raises a key question: is there a vast database to match the AI’s analytical capabilities? If the data exists, are there AI agents that can gather various data sources as needed, allowing developers to write prompts for different scenarios? These questions have sparked a broader discussion about the existing gaps and opportunities in the field of cryptocurrency and traditional industries in the data and AI sectors.

1. Data silos and monopolies: Although many data service companies, such as Dune Analytics, Nansen, Chainbase, and Alchemy, have indexed data from almost all blockchains, this data is still centralized. Data that could be indexed once and made universally available is repeatedly indexed, lacking interoperability between different entities, resulting in low data utilization efficiency in the industry. In addition, the off-chain data of platform X (formerly Twitter) is gradually being monopolized, violating the principles of decentralization and trustlessness. As a result, many data projects have been labeled as Web2.5, failing to fully embrace the native concepts of encryption.
2. Lack of unified data collection methods and standards: Although there are many open-source frameworks for processing on-chain data ETL, there is no consensus on the definition of data, resulting in large differences in analysis results between platforms. For example, the definition of “active user” on the blockchain can vary significantly depending on the criteria chosen by analysts, highlighting the subjective nature of data interpretation. This problem is exacerbated by the lack of tools for systematically collecting and organizing data from Platform X.
3. Unfairly valued data: Platforms like Dune and Footprint host community-created data analytics dashboards that centralize the data and insights generated. However, the analysts and developers who provide these insights often don’t reap the actual financial benefits, perhaps simply reputation. In addition, platform X’s high API usage fees and monopoly control of data hinder the openness of data. If Platform X restricts API access, it will effectively monopolize the data, forcing all users to bear high costs.
4. Most proxy projects ignore data needs: While many projects claim to build proxy platforms, proxy builds are based on massive amounts of data. Without data, these efforts will be in vain. Data sets the upper limit of AI capabilities, and AI determines the efficiency with which data is used. At this stage, it’s critical to build data barriers and tools, remembering that algorithms, computing power, and data are the immutable trinity of the AI industry.
5. AI agents as the next generation of assets: Comprehensive infrastructure including the data layer, AgentOps, service layer, and application layer is crucial for AI agents. Each layer holds tremendous opportunities, emphasizing the necessity of developing next-generation assets in the AI and cryptocurrency ecosystem.

3. The technical architecture of the DIN

The DIN blockchain architecture consists of four levels: consensus layer, data layer, service layer, and application layer, which gradually advance from basic technology to upstream applications.

• Consensus Layer: AI agents need to have credibility and transparency in decision-making and reasoning processes. The DIN blockchain provides decentralized security through its consensus layer, based on OP Stack and utilizing the security of the BNB chain. This layer ensures transparent and tamper-resistant operations and decisions, providing a reliable execution environment for AI agents.

• Data layer: AI agents need high-quality, scalable data for training and decision-making. The data layer of the DIN blockchain enables efficient processing of on-chain and off-chain data, facilitating the development of AI decentralized applications (dApps) and AI agents. A key component of this process is the DIN Chipper node, which plays a crucial bridging role in the data pre-processing process.

As a licenser of data validators and vectorizers, DIN Chipper nodes are responsible for validating, categorizing, cleaning, and enhancing raw data. These nodes are equipped with small language classification models trained using fastText, allowing for fast and accurate processing of multilingual texts. This capability ensures that the data is effectively prepared to meet the high standards for AI agents to perform in diverse applications.

• Service Layer: The service layer provides the necessary tools for creating and deploying AI agents. It includes LLMOps, a comprehensive framework covering the deployment, monitoring, and optimization of large language models, ensuring that AI agents can efficiently handle complex computing tasks. Through Prompt as a Service and Agentic Workflow, the service layer simplifies the process of creating and deploying AI agents, enabling them to efficiently execute complex tasks.

In addition, the service layer of the DIN blockchain supports multi-agent collaboration through RAG (Retrieval Augmented Generation) and Agentic Workflow. RAG provides efficient knowledge search, indexing, and retrieval, promoting data sharing and collaboration among different AI agents; Agentic Workflow offers a visual workflow interface to assist in constructing and managing complex tasks of multi-agent collaboration, supporting seamless cooperation among agents.

• Application Layer: The application layer of the DIN blockchain has demonstrated its strong potential through multiple self-developed decentralized applications (dApps), including Analytix, Reiki, and xData, which provide rich value for business use cases on the BNB Chain. Analytix is an on-chain data analytics platform trusted by major customers such as BNB Chain, Moonbeam, and Manta Network to help users create detailed dashboards to analyze on-chain activity. Combining AI with user-generated content (UGC) to provide creators with customized AI agency and dataset management capabilities, Reiki has been a notable success on Product Hunt, earning first place in Product of the Day and Product of the Month, and third place in the 2024 year-end rankings. At the same time, xData is a revolutionary AI data collection tool that has attracted more than 30 million registered users, and has partnered with AISpeech to provide multilingual voice data annotation services for the automotive industry using blockchain technology, further promoting its commercial application in the field of AI voice. These dApp achievements fully demonstrate the powerful capabilities and extensive influence of the DIN blockchain application layer.

DIN blockchain has natural advantages in providing AI agents with powerful data support, a trusted execution environment, efficient computing power, and multi-agent collaboration capabilities. In addition, the DIN blockchain promotes the active growth of the ecosystem by introducing the native token $DIN as a gas fee, while providing a low-cost operating environment for users and developers. Through economic incentives, the DIN blockchain provides a clear revenue path for developers and users, attracts more developers to participate, and promotes the widespread application of AI agent technology.

3.1 Data Layer: Everything for Data

“Every data scientist should spend 80% of their time on data preprocessing and 20% of their time on analysis.” As shown in Figure 1, data plays a crucial role in the entire AI workflow and is the foundation of AI blockchain - all dAI applications rely on high-quality data. We have designed and implemented the DIN protocol, which ensures that the network can obtain high-quality data by adopting incentive mechanisms. This chapter will explain how the DIN protocol works for network participants and how they can be rewarded by participating in data activities.

3.2 Service Layer: AI Agent Toolkit

At the service layer, a development platform is provided that combines the concepts of Backend-as-a-Service and LLMOps, enabling dAI application developers to quickly build production-grade generative applications. Even non-technical personnel can participate in AI application definition and data operations.

By integrating the key technology stacks needed to build LLM applications, including support for hundreds of models, an intuitive Prompt orchestration interface, a high-quality RAG engine, and a flexible Agent framework, with a suite of easy-to-use interfaces and APIs, the platform saves developers a lot of time and reinvents the wheel, allowing them to focus on innovation and business needs.

The focus of this layer is to define and continuously improve the development of dAI applications, with advantages including:

• Integrate LLM into existing business - enhance the capabilities of current applications by introducing LLM. Access ready-made RESTful APIs to decouple Prompts from business logic. The management interface tracks data, costs, and usage, while continuously improving performance.

• LLM infrastructure - serving as an internal LLM gateway to accelerate the adoption of GenAI technology.

• Explore the capabilities of LLM - easily perform Prompt engineering and Agent technology practices through different LLMs and integrate them with other real-time external knowledge.

3.3 Application Layer: Ecosystem and Products

DIN is a token-based incentive network that incentivizes users to participate through the entire data processing lifecycle. Users are rewarded for participating in a variety of activities, which drives active participation and fosters a vibrant community.

We have introduced Analyix and xData as dedicated data collection layers to enhance the data collection process. Analyix focuses on real-time structured on-chain data, providing valuable insights and information. xData, on the other hand, focuses on collecting off-chain data to ensure the comprehensiveness and diversity of the data pool.

Web3Go DIN prioritizes data quality through a rigorous verification mechanism. This mechanism transforms objective and subjective data into reliable and credible information. By validating data, Web3Go DIN ensures that the data consumed by applications and scenarios is of the highest quality.

One key application in the Web3Go DIN is Reiki, an innovative data consumption application. Reiki leverages advanced AI technology to unlock the potential of the collected data. Reiki generates valuable insights through its AI capabilities, offering a wide range of use cases in real-world scenarios.

With its token-based incentive networks, Analyix, xData, and Reiki, DIN will fundamentally change the way data is processed and consumed. By empowering users to participate and provide high-quality data, DIN opens up new possibilities for other innovative and transformative applications.

4. DIN Token Economics

4.1 DIN Token Distribution

The total supply of $DIN is capped at 100,000,000 tokens, allocated as follows:

1. Community

In order to incentivize contributors to various products on the DIN chain, including DIN Chipper nodes and xData, a portion of $DIN tokens will be distributed to the community:

• Initial unlock: 10% of the community allocation will be unlocked at TGE.

• Airdrop: Eligible participants will receive token airdrops on the Binance Smart Chain (BSC). Active 2nd to 10th level node holders will receive additional airdrop rewards.

• Future allocation: The remaining community tokens will be linearly allocated within 48 months to ensure consistent growth with DIN and long-term benefits.

2. Ecosystem

This allocation supports the sustainability and innovation of DIN development by funding the following projects:

• Plan: Research and development, marketing activities, on-chain development and maintenance, node operator management, and other projects on the DIN chain.

• Distribution:

◦ 15% of the ecosystem allocation will be unlocked at TGE.

◦ The remaining tokens will be linearly vested over 30 months.

3. Investors

To reward early supporters:

• Initial unlock: 5.47% of the allocation will be unlocked at TGE.

• Vesting schedule: A 24-month vesting plan with a cliff period of 3 months for the first tranche.

4. Team and Advisors

To reward the contributions of the DIN team and advisors:

• Vesting timeline: 6 months of cliff period, followed by 30 months of linear vesting.

5. Market making and liquidity

In order to support the liquidity and market making efforts of $DIN:

• Distribution: 5% of the total supply.

• Unlock: 100% of this allocation will be unlocked at TGE.

The DIN token will be listed on the spot trading area of Gate.io platform at 20:00:00 on 2025/02/14 (UTC+8), please stay tuned!

4.2 Application scenarios of DIN tokens in the ecosystem

Conclusion

DIN is a decentralized blockchain network designed to provide AI agents with reliable data support, a secure execution environment, and efficient computing power. DIN drives innovation in data and artificial intelligence (AI) by adopting advanced blockchain technology, AI capabilities, and a decentralized ecosystem.