1. Report Highlights

   1.1 Core Investment Logic

• Quilibrium aims to find a "balance" between the computing power of the traditional Internet and the decentralization of blockchain, and has designed a unique decentralized cloud computing architecture for this purpose.

• Quilibrium has built a database-based operating system that is more user-friendly in terms of development experience, which may attract more traditional software developers and facilitate current Web3 developers to build more complex encrypted applications.

• Quilibrium's design emphasizes security and privacy, making it attractive to enterprises that do not want to expose sensitive data but wish to use encryption technology; for individuals, Farcaster's initial breakthrough also proves the long-term potential of decentralized applications in acquiring users and generating revenue.

• Founder and CEO Cassie Heart is a former senior engineer at Coinbase and developer of Farcaster, and the team has rich experience, stable delivery capabilities, and distinctive personalities.

  1.2 Main Risks

• The project is in a very early stage, with the mainnet not yet launched, and the validation of technical feasibility and market demand has not been completed.

• In the short term, it may face competition from Arweave AO, which has higher visibility in terms of user awareness and developer support.

• There is no fixed token model, and the token release rate may be unstable, which increases certain risks for investors.

  1.3 Valuation

  Due to Quilibrium being in a very early stage, we are temporarily unable to determine the project's accurate valuation. However, from the perspective of circulating market value and total circulating market value, Quilibrium's current market value is relatively attractive compared to other market players with overlapping concepts.

1. Business Analysis

   Quilibrium positions itself as a "decentralized Internet layer protocol that provides the convenience of cloud computing without sacrificing privacy or scalability," as well as a "decentralized PaaS solution." In this section, the business of Quilibrium will be elaborated on around the following questions.

• What are the problems with traditional Internet cloud computing?

• Why do we need (another) decentralized computer?

• What makes Quilibrium special compared to the current mainstream blockchain designs?

  

  2.1 Business Positioning

  2.1.1 Starting from Computing

  Regardless of Web2 or Web3, "computing" is a crucial concept and the driving force behind application development, execution, and expansion.

  In traditional Internet architecture, computing tasks are usually completed by centralized servers. The emergence of cloud computing has improved the scalability, accessibility, and cost efficiency of computing, gradually replacing traditional computing as the mainstream.

  In terms of service content, the cloud service models provided by large cloud service providers can generally be divided into three categories: Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS), corresponding to different needs and levels of control over resources. End users are more familiar with SaaS, while PaaS and IaaS are mainly targeted at developers.

  

  

  In mainstream blockchains such as Ethereum, computing is usually completed by decentralized nodes. This method does not rely on centrally controlled servers; each node executes computing tasks locally and ensures data correctness and consistency through consensus mechanisms. However, the computing capabilities and processing speed of decentralized computing usually cannot compare with traditional cloud services.

  Quilibrium aims to find a "balance" between the computing capabilities and scalability of the traditional Internet, and the decentralization of blockchain, opening up new possibilities for application development.

  

  2.1.2 Centralization Issues of Computer Systems

  For most end users, the centralization issues of computer systems are not easily perceptible. This is because most end users mainly interact with the hardware aspects of computer systems. Our PCs, mobile phones, and other devices are distributed worldwide and run independently under personal control. This distributed physical existence makes computer systems not necessarily centralized at the hardware level.

  In contrast to the relatively dispersed hardware, existing computer systems are significantly more centralized at the network architecture and cloud computing service levels—Amazon AWS, Microsoft Azure, and Google Cloud held over 67% of the cloud service market share in the first quarter of 2024, creating a significant gap with later entrants.

  

  Furthermore, as a "water seller" in the AI wave, the trend of the strong getting stronger among cloud service providers seems to be continuing. Microsoft Azure, as the exclusive cloud service provider for OpenAI, has shown accelerated growth in performance over the past year. In Microsoft's third quarter of the 2024 fiscal year (the first quarter of 2024), the revenue from Azure and other cloud services grew by 31%, exceeding the market's expected growth of 28.6%.

  

  In addition to market competition considerations, the privacy and security issues brought about by centralized computer systems are increasingly receiving attention—each outage of several major cloud service providers has a widespread impact. Data shows that from 2010 to 2019, AWS experienced 22 sudden failures, with an average of 2.4 failures per year. In addition to the impact on Amazon's own e-commerce business, the network services of companies using AWS, such as Robinhood, Disney, Netflix, and Nintendo, also experienced large-scale interruptions.

  2.1.3 Proposal of Decentralized Computers

  Against this background, the necessity of decentralized computers has been repeatedly emphasized. As more and more centralized cloud service providers adopt distributed architectures to avoid single point failures by replicating data and services in multiple locations, and to improve performance through edge storage, the narrative of decentralized computing has gradually shifted to focus on data security, privacy, scalability, and cost-effectiveness.

  First, let's analyze the concepts of decentralized computers proposed by different projects. Their common feature is the hope to build a global distributed computing platform that supports the development of decentralized applications through decentralized data storage and processing.

• World Computer: Generally refers to Ethereum, which provides a global environment for smart contract execution. Its core function is decentralized computing and globally unified execution of smart contracts.

• Internet Computer: Generally refers to ICP developed by the Dfinity Foundation, aiming to extend the functionality of the Internet and enable decentralized applications to run directly on the Internet.

• Hyper Parallel Computer: Generally refers to the AO protocol proposed by Arweave, a distributed computing system running on the Arweave network, known for its high parallelism and fault tolerance.

It is worth noting that ICP, AO, and Quilibrium are not traditional blockchains. They do not rely on a linear block arrangement structure but maintain core principles of blockchain such as decentralization and data immutability, and can be considered as a natural extension within the scope of blockchain technology. Although ICP has not yet realized its grand vision, the emergence of AO and Quilibrium has indeed brought a new possibility that could impact the future of Web3.

The table below compares the technical characteristics and application directions of the three, aiming to help readers understand whether "Quilibrium will repeat the mistakes of ICP" and the differences between Quilibrium and AO, which is known as the "Ethereum killer" as both are cutting-edge solutions for decentralized computing.

2.2 Consensus Mechanism

In traditional blockchains, the consensus mechanism is at a more abstract and core level, defining how the network reaches consensus, processes and verifies transactions, and other operations. The choice of different consensus mechanisms will affect the network's security, speed, scalability, and degree of decentralization.

Quilibrium's consensus mechanism is called "Proof of Meaningful Work (PoMW)," where miners are required to perform meaningful work for the network, such as data storage, data retrieval, and network maintenance. The design of the PoMW consensus mechanism integrates multiple fields such as cryptography, multi-party computation, distributed systems, database architecture, and graph theory, aiming to reduce reliance on a single resource (such as energy or capital), ensure the decentralization of the network, and maintain security and scalability as the network scales.

The incentive mechanism is crucial to ensure the smooth operation of the consensus mechanism. Quilibrium's incentive allocation is not static but dynamically adjusts based on the network's state to ensure that incentives match demand. Quilibrium also introduces a multi-proof mechanism, allowing a node to verify multiple data fragments, even when nodes and core resources are insufficient to maintain network operation.

A simplified formula can be used to understand a miner's final earnings, where the unit reward dynamically adjusts based on the network's scale.

Earnings = Score × Unit Reward

The calculation of the score is based on various factors, with the specific formula as follows:

Where the parameters are defined as follows:

1. Time in Mesh for Topic: The longer the participation time and the higher stability, the higher the score.
2. First Message Deliveries for Topic: The more times messages are delivered for the first time, the higher the score.
3. Mesh Message Delivery Rate/Failures for Topic: Nodes with higher delivery rates and lower failure rates receive higher scores.
4. Invalid Messages for Topic: The fewer times invalid messages are delivered, the higher the score.

The weighted sum of these four parameters will have an upper limit (TC) for a specific topic, to avoid unfair scoring due to excessively large parameters.

5. Application-Specific Score: Score defined by a specific application.
6. IP Collocation Factor: The fewer nodes from the same IP address, the higher the score.

The number of nodes currently running on Quilibrium has exceeded 60,000, and actual node earnings may fluctuate based on the different parameter weights between each version. After version 1.4.19, miners can view their earnings in real-time, but they will need to wait for the mainnet to be launched to claim them.

2.3 Network Architecture

Quilibrium's core business is a decentralized PaaS solution, and its network architecture mainly consists of communication, storage, data querying and management, and operating systems. This section will focus on its design differences compared to the current mainstream blockchains, and readers interested in technical details and implementation methods can refer to the official documentation and whitepaper.

2.3.1 Communication

As the foundation of the network, Quilibrium's communication consists of four parts.

a. Key Generation

Quilibrium proposes a PCAS (Planted Clique Addressing Scheme) key generation method based on graph theory. Similar to traditional blockchain technology, PCAS also uses asymmetric encryption—each user has a public key and a private key, where the public key can be made public for encrypting information or verifying signatures, and the private key is kept confidential for decrypting information or generating signatures. The difference lies mainly in the key generation method, representation, and application direction (see the table below).

b. End-to-End Encryption

End-to-end encryption (E2EE) is a critical component for ensuring secure communication between nodes, where only the communicating parties can see plaintext data, and even systems or intermediaries helping to transmit information cannot read the content.

Quilibrium adopts a method called Triple-Ratchet for end-to-end encryption, which provides higher security compared to traditional ECDH schemes. Specifically, traditional schemes usually use a single static key or periodically update keys, while the Triple-Ratchet protocol updates keys after each communication, achieving forward secrecy, post-compromise security, plausible deniability, replay protection, and unordered message delivery. This method is particularly suitable for group communication, but it also comes with higher complexity and computational costs.

c. Mixnet Routing

Mixnets are a black box that can receive information from senders and pass it on to receivers, making it impossible for external attackers to associate senders and receivers even if they can access information outside the black box.

Quilibrium uses RPM (Random Permutation Matrix) technology, providing a mixnet architecture that is structurally complex and difficult for external and internal attackers to crack, with advantages in anonymity, security, and scalability.

d. Peer-to-Peer Communication

(Continued in next message)

GossipSub is a peer-to-peer message delivery protocol based on the publish/subscribe model, widely used in blockchain technology and decentralized applications (DApps). Quilibrium's BlossomSub protocol is an extension and improvement of the traditional GossipSub protocol, aiming to enhance privacy protection, resistance to Sybil attacks, and optimize network performance.

2.3.2 Storage

Most traditional blockchains use cryptographic hash functions as the basic tool for data integrity verification and rely on consensus mechanisms to ensure network consistency. Such mechanisms have two main limitations:

• They usually do not include verification of storage time, and there is no direct mechanism to defend against attacks based on time or computational power.
• Storage and consensus mechanisms are often separate, which may lead to issues with data synchronization and consistency.

Quilibrium's storage solution uses a Verifiable Delay Function (VDF) design to create a time-dependent chain structure and integrates storage and consensus mechanisms. Several characteristics of this approach can be summarized when combined with the diagram below:

• Input Processing: By using hash functions such as SHA256 and SHAKE128 to process input, any small changes in data will result in significantly different hash values, making the data more difficult to tamper with and easier to verify.
• Delay Guarantee: The computation process is intentionally set to be time-consuming. The computation tasks must be executed in sequence, with each step depending on the result of the previous step, making it impossible to speed up the process by increasing computational resources to ensure that the output is based on continuous and deterministic computation. Since the generation process cannot be parallelized, any attempt to recalculate or modify the published VDF result will require a considerable amount of time, providing enough time for network participants to detect and respond.
• Fast Verification: The time required to verify a VDF result is much less than the time required to generate that result, usually only requiring some mathematical checks on the final result or using some auxiliary data to confirm the validity of the result.

Source: Quilibrium Whitepaper

This time-based proof chain structure does not rely on block generation in traditional blockchains and theoretically can reduce MEV attacks and front-running.

2.3.3 Data Query and Management

Traditional blockchains mostly use simple key-value storage or Merkle Trees to manage data, which is often limited in expressing complex relationships and supporting advanced queries. Additionally, most blockchain systems currently do not provide built-in privacy protection mechanisms when nodes execute queries, providing the background for privacy-enhancing technologies such as zero-knowledge proofs.

Quilibrium proposes an "Oblivious Hypergraph" framework, combining hypergraph structure and Oblivious Transfer technology, to support complex query capabilities while maintaining data privacy:

• Hypergraph Structure: Allows edges to connect multiple vertices, enhancing the ability to express complex relationships. This structure can directly map various database models, allowing any type of data relationship to be expressed and queried on the hypergraph.
• Oblivious Transfer Technology: Even nodes processing data cannot know the specific content being accessed during data queries, enhancing privacy protection during the data query process.

2.3.4 Operating System

An operating system is not a native concept in blockchain. Most traditional blockchains mainly focus on consensus mechanisms and data immutability, usually not providing complex operating system-level functionality. For example, while Ethereum supports smart contracts, its operating system functionality is relatively simple, mainly limited to transaction processing and state management.

Quilibrium has designed an operating system based on its hypergraph database and implemented common operating system primitives, such as a file system, scheduler, IPC-like mechanisms, message queues, and control key management. This design of building an operating system directly on the database can provide support for developing complex decentralized applications.

Source: Quilibrium Whitepaper

2.4 Programming Language

Quilibrium's development primarily uses Go as the programming language, combined with Rust and JavaScript. The advantage of Go lies in its ability to handle concurrent tasks, concise syntax, and an active developer community. According to the Tiobe programming language rankings, Go has seen a significant rise in rankings in recent years and is currently ranked 7th in the latest June rankings. Other blockchain projects that also use Go for low-level development include Ethereum, Polygon, and Cosmos.

Source: Quilibrium

Source: Tiobe

3. Project Status

3.1 Project History and Roadmap

Quilibrium's whitepaper was released in December 2022, and its roadmap is roughly divided into 3 phases: Dusk, Equinox, and Event Horizon.

Quilibrium is currently in a very early stage, with the team iterating on network updates every two weeks. The latest version is v1.4.20. Since the team has removed the 1.5 phase from the roadmap, the network will directly upgrade to version 2.0 after version 1.4. The 2.0 version, also known as the mainnet, marks the end of the Dusk phase and is expected to be officially launched in late July, allowing for the bridging of $QUIL.

According to the tentative plan, the Equinox and Event Horizon phases will provide support for more advanced applications such as streaming media and AI/ML model training.

3.2 Team and Funding

The founder/CEO of Quilibrium is Cassie Heart. Prior to founding Quilibrium, she was a senior software engineer at Coinbase, with over 12 years of experience in software development and blockchain.

Due to Cassie's opposition to centralized social media platforms, she and the Quilibrium project account are mainly active on Farcaster. Cassie's Farcaster account has over 310,000 followers, including Ethereum founder Vitalik. Cassie is also a developer for Farcaster.

According to the developer data panel for the Quilibrium project, development began in April 2023 and has been consistently ongoing. There are a total of 24 developers, with Cassie Heart (Cassandra Heart) as the main contributor.

Source: Quilibrium

Quilibrium's team has not publicly disclosed its funding history and investment institutions.

3.3 Token Model Analysis

$QUIL is the native token of Quilibrium, launched in a 100% fair manner, with all token production coming from node operation. The team operates a small portion of nodes, but their token holdings account for less than 1%.

$QUIL does not have a fixed token model, and the total token supply is unlimited, but it will dynamically adjust based on the network's adoption speed. When the network grows, more tokens will be released as node incentives; if the growth slows down, the token release rate will also decrease accordingly.

The table below shows the predictions made by the team and community members regarding the token release schedule. The current circulating supply is 340 million, and the expected final supply will converge around 20 billion. The specific release situation will depend on the development of the ecosystem.

Source: @petejcrypto

3.4 Risks

Potential risks for Quilibrium at the current stage include:

• The project is in a very early stage, with the mainnet not yet released, and the project's complexity is high. The validation of technical feasibility and market demand has not been completed.
• In the short term, it may face competition from Arweave AO, which has higher visibility in terms of user mindset and developer aspects.
• The lack of a fixed token model may lead to unstable token release rates, increasing certain risks for investors.

4. Valuation

The valuation of infrastructure for quasi-public chains is a very complex process, involving multiple dimensions such as TVL, on-chain active addresses, dApp numbers, developer communities, etc. Since Quilibrium is still in a very early stage and the token $AO of Arweave AO has not yet been traded, we cannot currently determine the project's accurate valuation.

Here, we list the circulating market cap and fully diluted market cap of projects with some conceptual overlap with Quilibrium (data as of June 23, 2024) for reference.

Source: CoinGecko, data as of June 23, 2024

5. References and Acknowledgments

The writing of this article would like to thank Hai Ge (@PleaseCallMeWhy), Lan Ge, and Connor for their review and feedback.

• https://quilibrium.com/quilibrium.pdf
• https://paragraph.xyz/@quilibrium.com
• https://dashboard.quilibrium.com/
• https://www.youtube.com/watch?v=Ye677-FkgXE&ab_channel=CassandraHeart
• https://dune.com/cincauhangus/quilibrium
• https://source.quilibrium.com/quilibrium/ceremonyclient/-/graphs/main?ref_type=heads
• https://www.tiobe.com/tiobe-index/
• https://www.blocmates.com/meal-deal-research-reports/quilibrium-crypto-not-blockchain-long-live-the-internet
• https://www.statista.com/chart/18819/worldwide-market-share-of-leading-cloud-infrastructure-service-providers/
• https://s2-labs.com/admin-tutorials/cloud-service-models/
• https://medium.com/@permadao/%E5%8E%BB%E4%B8%AD%E5%BF%83%E5%8C%96%E4%BA%91%E6%9C%8D%E5%8A%A1%E8%BF%9B%E5%8C%96%E5%8F%B2-%E4%BB%8E-dfinity-ic-%E5%88%B0-arweave-ao-839b09b4f3ff
• https://www.microsoft.com/en-us/investor/earnings/FY-2024-Q3/press-release-webcast
• https://x.com/perma_daoCN/status/1798565157435830416
• https://x.com/Pow2wer/status/1802455254065402106

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。