CoinW Research Institute: In-depth Report on Al Agents
Key Points
● Traditional AI, while effective in many scenarios, suffers from limitations such as high centralization and high costs, which make them perform poorly in dynamic, decentralized, real-time decision-making environments. AI agents combined with cryptography can compensate for these shortcomings, These agents possess higher autonomy, adaptability, decentralization, and multitasking capabilities, playing a key role in complex economic models and real-time decision-making, thereby optimizing the efficiency of the entire AI system’s operation.
● The AI Agent framework is the infrastructure of AI Agent and an important development direction in the AI Agent track. Currently, representative projects include ai16z, zerebro, and arc. ai16z is based on the Eliza framework, which has been open-sourced on Github. Currently, the Eliza framework is widely used in various agent applications, including wallets, social integration, and customizable agents, making ai16z a representative project in the community for AI agent frameworks. The framework of zerebro is ZerePy, which is an open-source Python framework. Its initial function is mainly to motivate users’ participation on X, and it will expand to other platforms in the future and support integrated memory and local LLM. However, the product has not yet been launched and needs further observation. arc is currently the only project based on the RIG framework and is still in the early stage of development. Considering that the currently most widely used Eliza framework is a lightweight basic framework, there may be upgraded framework representative projects in the future.
● In the direction of AI Agent asset issuance, the representative ones are Virtuals Protocol and clanker on the Base chain, Dasha on the Solana chain, and AigentFi on the ZKsync chain. Virtuals focuses on the underlying foundation of entertainment and virtual agent issuance, with 12,069 existing ecosystem projects; clanker is positioned as the Pumpfun of the Base ecosystem, with 17,242 existing ecosystem projects; Dasha is positioned as the Pumpfun of the Solana ecosystem, with 2,401个 existing ecosystem projects; A(i)gentFi is based on the ZKsync chain, and its positioning
is the agent of all AI agents, with 55 existing ecosystem projects.As of the time of writing, the total market value of the Virtuals ecosystem token is 3 billion US dollars, with the top three tokens by market value being VIRTUAL at $2.4 billion, AIXBT at $230 million, and GAME at $90 million. The total market value of clanker ecosystem tokens is $110 million, with the top three tokens being CLANKER at $51.86 million, ANON at $18.75 million, and LUM at $12.18 million.Dasha and AigentFi are relatively lagging in development, with platform token market values of $50.22 million and $16.03 million, respectively.
● In the direction of the AI Agent engine, the representative ones are griffain on theSolana chain and Spectre AI on the ERC-20 chain. The core concept of griffain is to directly turn users’ ideas into actions, evolving from the read-write mode to the read-write-action mode. Griffain has also received official support from Solana, and with the help of agents, griffain has achieved assisting humans in shopping. GRIFFAIN was released on November 3, and its current market value is $270 million. Spectre AI is a chain-based search engine based on ERC-20, and its core mechanism is to provide users with comprehensive insights into encrypted data. Spectre AI was born in 2023 and transformed from an AI prediction robot to a chain-based search engine project in May 2024. Compared to Spectre AI, griffain has a slight edge in overall project narrative and ecosystem development.
●In the direction of AI Meme, AI Meme represented by GOAT is a new engine for the combination of AI and Crypto. AI Meme has become an indispensable part of the market. GOAT, TURBO, and BULLY are representative projects of AI Meme. As of the time of writing, Wintermute is still the third-largest holder of GOAT, holding 4.66% of the total supply. The ways of AI Meme are also gradually diversifying. Currently, there are projects that have attempted to generate memes/generate pictures/music, etc., but the market performance is obviously weak. In the direction of AI-generated Meme, more external traffic is needed.
●In addition to the above major directions, there are also some interesting observable directions. For example, the analytical AI agent aixbt and the comprehensive intelligent agent Simmi. Compared with aixbt, SIMMI’s main advantage is the breadth of its settings, not limited to specific fields; the combination of AI Agent and DeFi, AiFi Mode. The vision of Mode is to realize the interaction from AI to AI and expand the boundary of DeFi. Intelligent agents can independently execute complex financial management tasks; the AI Agent verification direction Seraph. Seraph is an AI Agent jointly released by the Virtuals and TAO subnet communities, and its main income source is the token income for verifying authenticity. The product has not been released yet; the AI Agent real-time API direction Creator.Bid. Creator.Bid has obtained the support of well-known VCs such as Mechanism Capital, Zee Prime Capital, and Moonrock Capital. On December 10, Creator.Bid was sold on FjordFoundry and was attacked by DDOS, but the community is still optimistic. Currently, Creator.Bid mainly operates in the form of points, and the next sale plan has not been announced.
●With the continuous emergence of AI Agents, infrastructure will become the market winner; as the infrastructure matures, more applications will shift from speculation to value landing; AI Meme has become an indispensable part of the market, AI Meme will drive the iteration of AI Agents; a single AI Agent is difficult to meet the needs of the market, and the cross-domain integration of AI Agents will become a necessity.
Table of Contents
Key Points …………………………………………………………………..….…2
Ⅰ.Cryptocurrency as a New Engine for AI Agents …………..…………..8
1. The Dilemma of the Traditional AI Market ……………………………9
2. Cryptocurrency Injecting New Vitality into AI Agent ……………..10
II. Overview of the Crypto AI Agent Track ……………………………….11
1. AI Agent Infrastructure Framework ……………………………………11
1.1 The Eliza Framework of ai16z …………………………………………..11
1.2 The ZerePy Framework of zerebro …………………………………….15
1.3 The RIG Framework of arc………………………………………………. 17
1.4 Summary ……………………………………………………………………..18
2. AI Agent Asset Issuance …………………………………………………….19
2.1 The Core Mechanism of Virtuals Protocol ……………………………19
In addition …………………………………………………………………………21
2.2 The Core Mechanism of clanker ………………………………………..23
2.3 The Core Mechanism of Dasha …………………………………………..24
2.4 The Core Mechanism of A(i)gent ………………………………………..26
2.5 Comparative Analysis ………………………………………………………26
3. AI Agent Engine ……………………………………………………………….28
3.1 The Core Mechanism of griffain …………………………………………28
3.2 The Core Mechanism of Spectre AI ……………………………………..30
4. AI Meme …………………………………………………………………………32
4.1 GOAT ……………………………………………………………………………32
4.2 TURBO ………………………………………………………………………….34
4.3 BULLY …………………………………………………………………………..34
5. Other Observational Projects ………………………………………………35
5.1 Analytical Intelligent Agent aixbt ……………………………………….35
5.2 Comprehensive Intelligent Agent Simmi ……………………………..36
5.3 The Combination of AI Agent and DeFi, AiFi ……………………..…37
5.4 AI Agent Verification DirectionSeraph ……………………….…..…..38
5.5 AI AgentReal-time API Creator.Bid …………………………………….39
III. Future Outlook ………………………………………………………………41
1. Infrastructure Becoming the Market Winner …………………………41
2. Applications Shifting from Speculation to Value Realization ……..41
3. AI Meme Driving the Iteration of AI Agent …………………………….42
4. Cross-field Integration Becoming a Necessity ………………………..42
Ⅰ. Cryptocurrency as a New Engine for AI Agents
The AI market is growing rapidly. According to research data from Mordor Intelligence, it is expected that by 2027, its market value will be between $780 billion and $990 billion, with a compound annual growth rate (CAGR) of 40% to 55%. Sam Altman has stated that 2025 will be the year when agents start working, and the era of AI Agents is approaching.
Traditional AI, while effective in many scenarios, suffers from limitations such as high centralization and high costs, which make them perform poorly in dynamic, decentralized, real-time decision-making environments. AI agents combined with cryptography can compensate for these shortcomings。 They have higher autonomy, adaptability, decentralization, and multitasking capabilities, and can play a key role in complex economic models and real-time decision-making, thereby optimizing the operational efficiency of the entire AI system.
1. The Dilemma of the Traditional AI Market
The cost of training models is rising. The continuous increase in the training cost of traditional AI models has hindered the evolution of AI. OpenAI’s GPT-4 used an estimated $78 million worth of computing resources for training, while Google’s Gemini Ultra had a computing cost of $191 million.
High centralization. Traditional AI technologies and platforms are monopolized by a few tech giants. These platforms limit the competition of other enterprises through technical barriers and data advantages, leading to market inequality. In addition, traditional AI systems usually rely on centralized data processing and decision-making execution, and all calculations and decisions are concentrated in a central server or system. This may lead to bottlenecks, inefficiencies, or excessive dependence on a single control point in some cases.
Furthermore, traditional AI exhibits poor model generalization capabilities. Traditional AI models perform excellently on specific tasks or datasets, but in practical applications, they may not adapt to different environments or data, and overfitting makes the generalization capabilities of traditional AI models relatively poor.
2. Cryptocurrency Injecting New Vitality into AI Agent
AI agents combined with cryptocurrency have a fairer incentive mechanism. They incentivize users to provide data, train models, or participate in tasks through the token economy. Users can directly participate in the data economy and get paid for their contributions, making the incentive system fairer.
AI Agents combined with cryptocurrency have stronger autonomy and intelligence, and can interact with multiple systems and services, offering greater scalability. For example, under the cryptographic framework, AI Agents can manage the purchase and sale of assets through digital wallets without traditional bank accounts, and can also track market fluctuations, economic indicators, user behavior, and multiple other factors in real-time, and coordinate them.
AI Agents combined with cryptocurrency can handle multitasking in complex environments. The distributed thinking, code thinking, and consensus thinking of cryptography bring new inspiration to AI Agents, enabling them to better handle complex, multi-variable tasks.
II. Overview of the Crypto AI Agent Track
1. AI Agent Infrastructure Framework
The AI Agent framework is the infrastructure of AI Agent and an important development direction of AI Agent. This report analyzes the most representative projects in the current framework direction, namely ai16z, zerebro, and arc.
1.1 The Eliza Framework of ai16z
@ai16zdao is the first project on daos.fun, founded by Marc Andreessen, the founder of the well-known VC a16z. The core mechanism of ai16z is to use AI agents to obtain market information on and off the chain, analyze community consensus, and automatically conduct token transactions. ai16z is based on the Eliza framework, which has been open-sourced on Github. Currently, the Eliza framework is widely used in various agent applications, including wallets, social integrations, and customizable agents. Therefore, ai16z is regarded as one of the representative projects of the AI agent framework by the community.
The technical framework of Eliza provides practical technical support for ai16z. The Eliza framework is a basic AI agent framework that supports the construction and deployment of AI agents on X, Telegram, and Discord. In addition, the Eliza framework has an innovative incentive mechanism built-in. 5% — 10% of the revenue of all projects developed on the Eliza framework will be allocated to the AI16Z DAO fund, although currently, it is only on a voluntary basis and not mandatory.Currently, this framework is in its infancy, and as various virtual agents are integrated, ai16z, as the underlying framework, will continue to optimize. Currently, based on ai16z, some very interesting agents have emerged and are constantly integrating into human life. For example, on December 12, the agent @ropAIrito created by the creator @ropirito helped a human order the first pizza.
ai16z has also initiated a new trend of AI DAO. The decentralized token issuance mechanism, AI-driven intelligent governance, and the openness and transparency of fundraising have once again opened up a new narrative for AI. Looking at the entire track, ai16z is a representative project of the AI infrastructure framework and a new trend in AI DAOs.
Throughout its development, ai16z has experienced several fluctuations in community sentiment, mainly related to the core community figure Shaw, such as the previous battle between the capitalization of Eliza.This also reveals the hidden dangers of the daos.fun mechanism. For example, if the fund manager engage in improper trading practices, the community is powerless. In addition, it is worth noting that currently, the market value and managed assets of most funds on daos.fun are seriously inconsistent. Taking ai16z as an example, the current market value is $830 million, but the assets managed are only $20.18 million, indicating a significant gap between market value and managed assets. If ai16z can eliminate existing risks and focus on ecosystem construction and technical updates, it will inevitably present a new perspective.
With the rise of ai16z, another closely related project, degenai, has also attracted attention. Degenai is known in the community as the first trust distribution system for AI and finance projects. degenai and ai16z are both created by the same developer, @shawmakesmagic. degenai was born in pumpfun, and its core function is to provide Alpha information on Twitter and conduct transactions using this Alpha information. degenai is the first official AI agent created by the ai16z team and is also the only token included in the white list of the ai16z fund. Currently, the developer of degenai has sent 4.2% of the tokens to the DAO of ai16z, and 8% of the future trading income of the AI robot of ai16z will be used to repurchase degenai tokens, which will have a flywheel effect with the ai16z ecosystem.
1.2 The ZerePy Framework of zerebro
The core narrative of zerebro is AI intelligent autonomous agents, which can achieve autonomous token issuance and promotion, and even play with music, NFTs, etc. In addition, the long-term goal of zerebro is to become an open-source framework, allowing users to start AI agents by simply accessing the API without writing code. According to the plan of its co-founder Jeffy, the framework of zerebro is called ZerePy. ZerePy is an open-source Python framework. Its initial function is mainly to motivate users’ participation on X, and it will expand to other platforms in the future and support integrated memory and local LLM. ZerePy will allow users to deploy their own agents on the X platform with the support of large language models (LLM) such as OpenAI or Anthropic.
Zerebro’s tokens are autonomously issued and disseminated by AI Agents. zerebro has already begun trials beyond text, making progress in various areas such as music and NFTs. In terms of music, the album “Lost In Transmission” released by zerebro has nearly 20,000 plays on Spotify. zerebro has also released a music collection and is actively promoting cooperation with artists. In terms of NFT, according to Opensea data, the total transaction volume of the genesis series NFT of zerebro is 62 ETH.
The Zerebro model is also constantly being optimized. According to the news from the co-founder Jeffy, Zerebro will launch a test version in the next two to three weeks. This test version mainly focuses on the open-source framework and has new progress in the verification node network and NFT cross-chain integration.
Overall, Zerebro is a project that dares to actively innovate. Jeffy dares to break the convention and has his own unique marketing strategy. He also has a grand vision for the planning of Zerebro. In addition to the combination of NFT and music gameplay, Jeffy plans to enter the fields of games and board games in the future. If the plans such as the verification node network and ZerePy can be put into practical application, Zerebro will have more possibilities.
1.3 The RIG Framework of arc
arc is currently the only project based on the RIG framework. arc is created by @Playgrounds0x, the founder of the RIG framework. RIG is a framework based on the Rust language and is especially suitable for AI applications that require high optimization and high performance. The core goal of arc is to build a decentralized ecosystem based on AI Agents and is committed to developing into a similar or even surpassing the huge ecological cluster of a16z.
arc has a clear plan for the function of the token, mainly to support the development of the Rig framework and the future system, as well as the experiments of arc, such as the operation on TG and X. In terms of token distribution, 90% is for circulation supply, 5.5% is allocated to the reward pool and treasury, and 4.5% is allocated to the team. Additionally, arc has disclosed the nine custodial wallet accounts of the team on its official website.
In the official introduction of arc, one can see references to the “red pill” and the “blue pill,” originating from the movie “The Matrix. In “The Matrix”, there is a choice between the red pill and the blue pill. The red pill represents an uncertain future, that is, facing the cruel truth, awakening, and fighting for change; the blue pill represents a beautiful prison, that is, accepting the existing reality and maintaining ignorant happiness. The “red pill” and “blue pill” are a popular meme culture, and arc officially chooses two themes, representing the inclusiveness and influence of the arc community’s meme culture to the greatest extent.
Compared with ai16z and Zerebro, arc is still in a very early stage. The official first tweet was sent on November 19. As of the time of writing, the Rig framework of arc has 1,000 stars on github, while ai16z currently has more than 4,000 stars.
1.4 Summary
Infrastructure frameworks are an important direction for the AI Agent track. Looking at the current representative projects, they are all based on different frameworks, but their overall goal is to create a comprehensive ecosystem of AI Agents. Comparatively speaking, ai16z’s Eliza framework has been more widely applied, and ai16z is also leading in terms of community size, project ecosystem, and community sentiment. According to Github data, the Eliza framework of ai16z ranked first in the monthly trend ranking in November. As a new project in the framework, although the code adoption rate of arc is far lower than that of ai16z, it proposes a unique framework and the concept of combining token economics with Meme ecosystem development, which is still worthy of attention.
From the perspective of the entire AI Agent infrastructure framework direction, with the emergence of more AI Agents, the underlying framework will also be gradually optimized. Considering that the current Eliza framework is a lightweight basic framework, in the future, there may be more new projects emerging.
2. AI Agent Asset Issuance
In the direction of AI Agent asset issuance, the representative ones are Virtuals Protocol and clanker on the Base chain, Dasha on the Solana chain, and AigentFi on the ZKsync chain.
2.1 The Core Mechanism of Virtuals Protocol
Virtuals Protocol is an IAO platform on the Base chain, that is, an initial AI agent offering platform. The core mechanism of Virtuals is a platform serving AI Agents, focusing on games and entertainment, and realizing shared benefits for participating users.
The current main income sources of Virtuals are the income from its tokens and corresponding service fees. Agents running on Virtuals and creating corresponding liquidity pools must purchase Virtuals’ platform token, VIRTUAL.
The total supply of VIRTUAL tokens is 1,000,000,000, with 60% allocated for public distribution, 5% for liquidity pools, and 35% for ecosystem development. At the same time, the income generated by agents within the Virtuals ecosystem is used to directly repurchase and burn on-chain, forming a deflationary mechanism. Virtuals Protocol also supports multiple integrated platforms, linking Web2 and Web3 users. For example, the virtual agent LUNA can live stream on Tiktok 24/7. Currently, LUNA has paused live streaming on Tiktok and is gradually shifting its product operation direction to Twitter. It is worth noting that LUNA has also attracted the attention of the leading VC Grayscale, and Grayscale has analyzed AI Agent using LUNA as a case.
In addition, the token issuance mechanism of Virtuals combines the design of the internal and external markets. Only when the number of VIRTUAL in the internal market reaches 42,420 will the external market be opened. It is worth mentioning that some users in the community have found that Virtuals has borrowed the logic of Uniswap in the internal market code structure. After that, the internal market tokens will be destroyed and automatically converted into external market tokens, directly flowing into the market circulation. This mechanism reflects the deep support of the Virtuals platform for AI Agents and constructs an ecosystem that is highly attractive for both application and speculation.
Virtuals is also considered a cash cow product among AI Agents. As of the time of writing, Virtuals has more than 12,069 Agents and has accumulated $38.16 million in revenue. According to Coingecko data, as of the time of writing, the total market value of Virtuals ecosystem tokens is $3 billion, with the top three tokens by market value being VIRTUAL at $2.4 billion, AIXBT at $230 million, and GAME at $90 million Recently, Virtuals has also initiated a large number of ecosystem token airdrops, with wallets that have interacted with Virtuals receiving airdrops of SERAPH, PH, and other ecosystem tokens, and the more transactions, the greater the value of the airdrops. Virtuals Protocol aims to create a virtuous cycle system where the ecosystem continues to develop, the market value of agents is higher, and the market value of VIRTUAL is higher, as shown in the figure below:
The Virtuals team, with a background in gaming guilds, has years of practical experience in gaming and entertainment and continues to upgrade its products. On December 11, Virtuals launched the Agent sandbox, a simulation experiment platform for AI agents. The Agent sandbox is a simulation experiment platform that allows developers to fully control the personality and goal setting of AI agents and enhance the agent’s capabilities by creating custom functions. On December 13, Virtuals Protocol accessed the infrastructure of Hyperbolic to provide basic services such as APIs.
Virtuals is not limited to the AI Agent Launchpad track and is also making efforts in multiple directions. Many comprehensive tools also support the Virtuals ecosystem. For example, BananaGunBot now provides tools for snapping up pre-binding and post-binding transactions of Virtuals agent tokens. Virtuals has also proposed the GAME framework. Based on the GAME framework, future agents can achieve interoperability between agents, such as realizing human-like interaction betweenLUNA and another virtual agent.
As the foundation of AI agents, backed by the resources of the Base chain, Virtuals has become the basis for an increasing number of AI Agents. Virtuals focuses on the large market value track of games and entertainment of AI Agents and is also constantly developing towards a comprehensive AI foundation. The team also has rich experience. If the project ecosystem can be solidly developed, there will be broader development prospects.
2.2 The Core Mechanism of clanker
The main function of clanker is to help users on the social platform Farcaster easily deploy ERC-20 standard tokens. Without requiring users to have any complex technical knowledge, clanker can automatically complete token creation, liquidity pool configuration, and liquidity lock-up operations. clanker and pumpfun also have differentiated approaches. Firstly, in terms of user groups, there are no restrictions on the token issuance group of pumpfun, while clanker is only aimed at Farcaster users; secondly, the pumpfun platform mainly focuses on asset issuance, while clanker focuses on transaction tax after token issuance, relying on asset transactions.
The main innovations of clanker are as follows: one-click token issuance without programming; 40% of the transaction handling fee is rebated to the community; automated token deployment, directly linking to Uniswap V3, etc. As of the time of writing, clanker’s ecosystem has a total of 17,242 projects, with a total market value of clanker ecosystem tokens at $110 million. The top three tokens by market value are CLANKER at $51.86 million, ANON at $18.75 million, and LUM at $12.18 million, with income and expenditure currently roughly balanced.
2.3 The Core Mechanism of Dasha
@vvaifudotfun is developed based on ai16z’s Eliza framework and also integrates Zerebro, known as the pumpfun of the Solana chain. Its core mechanism is to achieve threshold-free one-click token issuance. Similar to Virtuals, Dasha also requires 1,000 native tokens VVAIFU when creating an agent. The issuance of new AI Agents and the unlocking of new functions on the Dasha platform all require VVAIFU. At the same time, the official will repurchase VVAIFU according to the profit situation. The reason why Dasha has won market attention in a short time is mainly due to the recognition of Solana’s founder, @aeyakovenko.
Dasha has shifted from a static chatbot to a dynamic, autonomous system, and its main innovation is the innovation of the token economy mechanism. As long as users activate new AI agents or unlock advanced platform features (including Twitter/Discord/TG interactions and on-chain wallet/TikTok interactions, etc.), VVAIFU tokens are destroyed, maintaining a deflationary mechanism for the tokens.
On December 15, Dasha announced that it had aggregated the frameworks of AI16z, Zerepy, and arcdotfun. According to Dune data, as of the time of writing, the total number of agents created by VVAIFU is 2,401, with more than 250,000 tweets, nearly 100,000 messages in Telegram and private chats, and the token VVAIFU’s market value is $50.22 million.
2.4 The Core Mechanism of A(i)gent
AigentFi is the first AI Agent on the ZKsync, which is also the first expansion of AI Agent to Layer2. A(i)gentFi is a one-stop shop on the ZKsync, dedicated to creating and developing AI agents, and an autonomous AI entity capable of performing complex tasks such as transactions, market analysis, and content creation. AigentFi has ambitious prospects, positioning itself as the agent for all AI agents.
A(i)gentFi focuses on practicality, scalability, and efficiency, and its main advantages lie in DeFi automation, revenue sharing, and user AI autonomy. To create an AI agent on A(i)gentFi, 20 HOLD and corresponding handling fees are also required. The main income sources of A(i)gentFi mainly include token issuance fees, migration fees, and transaction handling fees.
On December 12, A(i)gentFi cooperated with the DeFi wallet Holdstation based on ZKsync to launch an activity where staking HOLD can obtain platform income. According to official data, A(i)gentFi currently has more than 360,000 active users, $38 billion in transaction volume, and $3.1 million in revenue fees, ranking at the top on the zkSync chain. There are currently a total of 55 ecosystem projects, and the token HOLD has a market value of $16.03 million.
2.5 Comparative Analysis
Although the above projects all belong to the AI Agent asset issuance platform, there are also some significant differences between them:
● Market Positioning: Virtuals Protocol is based on the Base chain and focuses on the underlying framework of entertainment and virtual agent issuance; clanker is positioned as the Pumpfun of the Base ecosystem; Dasha is positioned as the Pumpfun of the Solana ecosystem; A(i)gentFi is based on ZKsync and is positioned as the agent of all AI agents.
● Income Sources: The income source of Virtuals Protocol is the purchase and application of VIRTUAL tokens and corresponding service fees; the income source of clanker is mainly transaction handling fees, with a 40% rebate to users; the main income source of Dasha is agency service fees and token income; the main income sources of A(i)gentFi are mainly token issuance fees, migration fees, and transaction handling fees.
● User Groups: users of Virtuals Protocol are those who are deeply involved in the field of entertainment and virtual agents, clanker targets Farcaster users, Dasha targets ordinary users, developers, content creators, and A(i)gentFi targets Layer2 and DeFi users.
● Virtuals Protocol has 12,069 ecosystem projects, clanker has 17,242, Dasha has 2,401, and A(i)gentFi has a total of 55 ecosystem projects.
Overall, Virtuals has chosen the large field of entertainment and virtual agents in terms of market positioning, with more room for market value growth. clanker’s main source of income is transaction fees, with a high rebate. However, clanker’s user group is limited to Farcaster users only. It is worth noting that during the writing of this report, the growth rate of the number of agents in the clanker ecosystem is the fastest, with a weekly growth rate of about 30%. Dasha’s ecosystem development is relatively lagging, and A(i)gentFi’s ecosystem is still in its infancy. In addition, AI Agent asset issuance is essentially a competition of on-chain liquidity and market attention. In terms of issuance time and ecosystem maturity, Virtuals occupies the leading position.
3. AI Agent Engine
In the direction of the AI Agent engine, the representative ones are griffain on the Solana chain and Spectre AI on the ERC-20 chain.
3.1 The Core Mechanism of griffain
griffain is called the first on-chain encrypted ChatGPT of Solana by the co-founder of Solana, @aeyakovenko. griffain is also a hackathon project of Solana AI agents, and its developer @tonyplasencia3 is a core developer of Solana. The official Solana team strongly supports griffain.
The core concept of griffain is to directly turn users’ ideas into actions, evolving from the read-write mode to the read-write-action mode. The main issue it addresses is that current AI Agents are mostly stuck in the process of information gathering to information feedback, while griffain implements users’ concepts based on datasets, providing practical solutions without requiring additional user processing.
griffain’s sources of revenue are diverse, with core income coming from intelligent agent services, transaction fees, and subscription fees. griffain has also integrated NFT gameplay, and the distribution of tokens for holding genesis NFTs has been completed. In the early stages of launch, users could easily create agents on griffain with just 1 SOL.
griffain has transformed AI from a passive tool into an active agent, which can easily complete tasks such as staking, trading, and shopping, broadening the automation boundary of Web3. Similar to ai16z, griffain has achieved assisting humans in shopping. Currently, griffain, through agents like Baxus, has been able to help humans purchase items like whiskey and shorts, achieving a combination of on-chain and off-chain.
griffain simplifies the user experience and, according to official plans, can also provide customized services, which already have some practical applications. GRIFFAIN was released on November 3, and its current market value is $270 million. If griffain can continue to expand similar agent services in more areas, it will help further promote its user growth and ecosystem maturity, and is expected to become Solana’s on-chain ChatGPT and app store.
As of the time of writing, griffain has exceeded 2,000 ecosystem projects, and the following 5 agents are showcased by the official team: Agent Baxus, Agent Blink, Agent onchain Holiday, Agent sniper, and Agent Flipper.
3.2 The Core Mechanism of Spectre AI
Spectre AI is a chain-based search engine based on ERC-20, and its core mechanism is to provide users with comprehensive insights into encrypted data. Spectre AI was born in 2023 and transformed from an AI prediction robot to a chain-based search engine project in May 2024.
Spectre AI is currently in the closed beta stage. Using machine learning, Spectre AI provides detailed sentiment analysis, technical analysis, and token insights. The partners of Spectre AI involve Web2 and Web3, and the Web2 partners include leading enterprises such as Google and NVIDIA.
The total supply of the token SPECTRE is 10 million, with 5% used for transaction fees and 3% for locked wallets. The token is mainly used for product research and development, marketing, etc. The project has currently reached phase 3.2, and X Bubbles (a bubble chart reflecting Twitter sentiment) will be released on December 20.
Compared to Spectre AI, griffain has a slight edge in overall project narrative and ecosystem development. griffain not only has the official endorsement of Solana but also has the strong support of DEV Toly. We look forward to more landing applications of griffain.
4. AI Meme
AI Memes, represented by GOAT, are a new engine for the combination of AI and Crypto, and have become an indispensable part of the market. This report takes the most representative GOAT, TURBO, and BULLY as examples.
4.1 GOAT
The core narrative of GOAT is the AI Bot Terminal of Truth (short for ToT), which, after nearly three months of exploration, has achieved token issuance and autonomous marketing.The most interesting thing is that this AI Agent has already possessed human attributes and even convinced Marc Andreessen (the founder of A16Z) to donate to it. This led to a16z publishing a discussion on YouTube titled “TRUTH TERMINAL: How An AI Bot Became a Crypto Millionaire Millionaire.” It is noteworthy that GOAT’s market value, from breaking through $100million to surpassing $500 million, was significantly propelled by the actions of A16Z. As of the time of writing, GOAT’s market value stands at $540 million. Recently, ToT has also begun posting on Sina Weibo in Chinese. From the published roadmap, GOAT’s long-term goal is the personification of AI agents, possessing both physical and legal attributes that humans have both externally and internally.
Looking at its development trajectory, GOAT evolved from early experimental exploration, to community dissemination, and finally became a cryptocurrency with a high market value. In addition to the operators behind the scenes, the core of its success is the prominent role of the well-known investment institution a16z and the subsequent community spread. As of the time of writing, Wintermute remains the third-largest holder of GOAT, holding 4.66% of the total supply. Additionally, according to sentientmarketcap data, the total market value of AI Agent crypto is $5.2 billion, with GOAT accounting for 15.23% of the total market value.
4.2 TURBO
In fact, the earliest token generated by AI Memes is Turbo. TURBO is an ERC-20-based token created by ChatGPT-4 in April 2023, and its current market value is $560 million. Turbo has a very strong community foundation, which is in line with the attributes of Memes. However, Turbo’s market performance only developed after GOAT became explosive, not at the beginning of ChatGPT’s token issuance. It is evident that for AI Memes to enter the public eye, a community foundation and market environment triggers are needed.
4.3 BULLY
BULLY is a Meme token related to @dolos_diary, emphasizing the sense of humor and active community interaction in the cryptocurrency field. It crosses the theme of overcoming challenges, presented in a light and symbolic manner. Its core mechanism is the automatic destruction mechanism in the smart contract, integrating the token economy into the entire ecosystem including on-chain games, NFT collection and trading, and social interaction.
The ways of AI Meme are also gradually diversifying. Currently, there are projects that have attempted to generate memes/generate pictures/music, etc., but the market performance is obviously weak. From the narrative of AI-generated Memes, entering the public eye requires the injection of external traffic, such as the support of well-known IPs from industry leaders/institutions/projects, and the widespread dissemination of a strong community. If it can trigger secondary dissemination in the community, coupled with the support of well-known market makers and other public elements, it will undoubtedly push AI Memes to new heights.
5. Other Observational Projects
5.1 Analytical Intelligent Agent aixbt
The core mechanism of aixbt is an AI intelligent agent that aggregates cryptocurrency hotspot information, providing users with market analysis and trading strategies. aixbt has attracted a significant amount of attention in a short period, not only being the largest market value project in the Virtuals ecosystem but also noteworthy for the holdings of the well-known market maker Wintermute. Currently, Wintermute holds approximately $1.31 million worth of AIXBT. As of the time of writing, from the on-chain address, Wintermute has no other transfer or other related actions except for holding.
aixbt is still in a relatively early stage, with a simple official website, and from the technical framework adopted, it shares a common framework with other projects, with no particular distinction. Currently, the known aggregation of aixbt is the aggregation of the remarks of several hundred KOLs on Twitter, and the official has no more content updates.
5.2 Comprehensive Intelligent Agent Simmi
@Simmi_IO and @EmpyrealSDK are two interconnected projects. Simmi is a part of the Empyreal project and provides services through its official AI agent Simulacrum. Empyreal uses the AI agent of Simulacrum to start on-chain intelligent agents from social media platforms through natural language processing technology. According to official EMP data, SIMMI’s weekly income has reached $600,000 through service fees and more. In terms of future planning, Simmi will be able to autonomously write roadmaps, publish whitepapers, hire collaborators, and perform a variety of comprehensive AI agent tasks.
What is worth noting about SIMMI is its native integration of Claude and ChatGPT, translating user intentions into on-chain operations. In this way, SIMMI can perfectly combine social media and on-chain interaction and accurately execute users’ intentions. According to Spot On Chain data, Wintermute also holds $377 worth of EMP, but as of the time of writing, there are no transfer or other related actions on this address.
Compared with aixbt, the main advantage of SIMMI is the wide range of its setting fields, not limited to specific fields. In the SIMMI plan, office automation, entertainment, and education automation are all its expandable market spaces. The AI Agent intelligent agent is currently still in the initial exploration stage, and more products that solve human actual needs will emerge in the future, which may involve all aspects of human work and life.
5.3 The Combination of AI Agent and DeFi, AiFi
AiFi is the product of the combination of AI Agent and DeFi and is also a new wave of DeFi automation. AiFi hopes to achieve seamless interaction of artificial intelligence agents on the chain, performing tasks such as MEV search, arbitrage, and smart contract auditing. It aims to achieve scalable, efficient, and non-custodial financial services, which is the full automation of AI Agents on the chain.
The current representative project of AiFi is Mode. The vision of Mode is to realize the interaction from artificial intelligence to artificial intelligence (AI2AI) and expand the boundary of DeFi. Intelligent agents can autonomously execute complex financial management tasks.In the latest project dynamics, it can be seen that the test subnet Synth of Mode has been launched. The uniqueness of the Synth test network lies in its access to the Bittensor (TAO) network and to Pyth’s price feeds, thereby stimulating the predictive and analytical capabilities of AI agents.
Compared with other AI agent projects, Mode was established in 2023 and has a relatively long development cycle. Currently, the only attempt in DeFi is to stake the token MODE to obtain OP rewards and other similar staking rewards in the DeFi field, and no more cases of actual integration with AI have been seen. After three airdrops, Mode currently also shows average performance in community activity.
5.4 AI Agent Verification DirectionSeraph
The verification of AI Agents is to solve the need to distinguish between true and false AIs as AI Agents rise. This includes the authenticity of agents and the authenticity of the content generated by AI agents. Seraph AI is currently a representative project in the direction of AI Agent verification.
Seraph is an AI Agent jointly released by Virtuals and the TAO subnet community, and its main source of income is the income from verifying authenticity. The product has not yet been released, and future product plans include links between AI agents and rewards for staking SERAPH in TAO. However, as for how to specifically verify, it currently only stays on the framework of Virtuals and with the help of the subnet of Bittensor, and the official has no more explanations. Currently, the white paper of Seraph is also a concise version published on github. Seraph has completed an airdrop on October 3, mainly targeting users of virtuals wallets and holders of the tokens Music and StTAO.
It is rather interesting that Seraph is also the first investment project of the on-chain VC sekoia_virtuals. In addition, sekoia has also analyzed the logic of its investment in Seraph.It posted on X that Seraph fills the current market gap and is an indispensable part of the market. The two also often interact on X, but currently, from the community’s heat, Seraph’s performance is average. Perhaps with a real product, Seraph may have some new perspectives.
5.5 AI AgentReal-time API Creator.Bid
Creator.Bid was established in 2023. @CreatorBid provides users with social media role agents connected to the real-time API of X and Farcaster.Any brand can start knowledge-based agents to execute content consistent with the brand on social platforms. At the same time, Creator.Bid also has plans for an AI Agent Launchpad. It is worth noting that Creator.Bid has obtained the support of well-known VCs such as Mechanism Capital, Zee Prime Capital, and Moonrock Capital.
Creator.Bid also has a high market attention, and it is interesting that Creator.Bid was attacked by DDOS during the sale on FjordFoundry on December 10, but there is still strong support from many community members. Currently, Creator.Bid mainly operates the ecosystem in the form of points, and the specific plan for the next sale has not yet been announced.
III. Future Outlook
Although AI Agent has made some progress, there are still some limitations in its development. The encrypted distributed thinking, code thinking, and consensus thinking of the blockchain bring new solutions. This report believes that with the continuous emergence of AI Agents, infrastructure will become the market winner; with the maturity of infrastructure, more applications will shift from speculation to value realization; AI Meme has become an indispensable part of the market, and AI Meme will drive the iteration of AI Agents; a single AI Agent can hardly meet the market demand, and cross-field integration will become a necessity.
1. Infrastructure Becoming the Market Winner
Looking at the development process of the industry, whether it is DeFi or NFT and other new engines of the market, it is inseparable from the construction of infrastructure. If AI Agent is the new engine of this bull market, then the construction of infrastructure will become the winner in the market. Infrastructure projects can accommodate the new traffic of AI Agent, provide support for technological innovation and implementation, and help achieve more intelligent and efficient decentralized applications. At the same time, more upgraded infrastructure projects will emerge, and we look forward to more landing applications.
2. Applications Shifting from Speculation to Value Realization
The current application of AI Agent is still in the early stage, and many lack the support of stable business models and actual user application scenarios. Taking the asset issuance platform as an example, more AI Agents still stay in the narrative of Meme, with strong speculative attributes. Platforms with real practical value will gradually establish mechanisms such as reliable custody and high-quality asset screening to promote the value realization of AI Agent.
3. AI Meme Driving the Iteration of AI Agent
AI Meme has become an important part of the market. AI Meme provides a unique driving force for the iteration of the technology and products of AI Agent in terms of dissemination, cultural promotion, user participation, and rapid feedback. By reducing the technical threshold, arousing extensive user interest, and encouraging collective creation, AI Meme not only provides an innovative iteration path for AI Agent but also accelerates its cross-industry and cross-field application and popularization. This cultural phenomenon is becoming an irresistible driving force for the technological innovation of AI Agent.
4. Cross-field Integration Becoming a Necessity
A single AI Agent can no longer meet the growing market demand. Different business scenarios and user needs prompt AI Agent to be continuously refined and diversified in fields such as finance, trading, and asset management. AI agents will not only be executors of automated tasks but also be able to achieve cross-field integration. For example, achieving multi-chain asset management, integrating smart contracts with traditional financial systems, cross-platform data integration and analysis, cross-industry applications, etc., thereby promoting the digital transformation and intelligent process of more industries and application scenarios.
