Author: Xiyou, ChainCatcher
Editor: Nianqing, ChainCatcher
In the past two months, the explosion of Agentfi has once again showcased the immense potential of the fusion of 'AI + Crypto' (also known as crypto AI). In a recent forecast of 2025 trends in the crypto industry, industry-leading institutions like a16z partners, Messari, and Blockworks all mentioned that the combination of 'AI and Crypto' will be a mainstream hotspot in the crypto market in 2025, with the intersection of crypto technology and AI expected to achieve breakthrough progress, driving the rise of numerous AI projects with tokens.
Looking back at 2024, the crypto AI market experienced unprecedented investment fervor, with major investment institutions rushing in and investment amounts skyrocketing. In this field, top venture capital institutions in the crypto industry, including Grayscale, Coinbase Ventures, Binance Labs, and a16z, have all actively engaged in the layout of 'Crypto + AI' projects.
Particularly, institutions like Delphi Venture, CoinFund, Coinbase Ventures, Binance Labs, and a16z have not only called for investments but also taken action, showcasing their strong optimism for the crypto AI field through multiple investments.
On December 24, Messari released the '2025 Crypto Outlook Report', indicating that crypto AI is the most striking emerging investment theme of 2024, with total financing in this field achieving approximately 100% year-on-year growth and the number of financing rounds increasing significantly by 138%. This article will systematically outline the layout of top crypto VCs in the Crypto + AI field in 2024, as well as the fundamentals of the specific projects they invested in.
Grayscale, Delphi, Coinbase, Binance Labs, and other crypto investment institutions have all laid out plans for 'Crypto + AI'.
Capital, as the core force driving industry progress and capturing cutting-edge projects, has always been seen as an important barometer for insights into market trends, based on its funding flows and public investment activities. The crypto AI market also welcomed an investment boom in 2024, with major investment institutions competing to enter, showing unprecedented enthusiasm for investing in crypto AI projects, and investment amounts are rapidly climbing.
Many top crypto venture capital institutions in the industry, such as Grayscale, Delphi Venture, Coinbase Ventures, Binance Labs, and a16z, have all boldly announced their active engagement in the investment layout and strategic deployment of 'Crypto + AI' projects.
At the beginning of this year (January), Tommy, co-founder of Delphi Digital, publicly stated on the social media platform X that he is highly optimistic about the integration of crypto and AI, encouraging related project developers to reach out. He also revealed that Delphi Ventures has invested in multiple crypto AI projects, including the decentralized GPU project io.net, AI blockchain OG Labs, and the AI venture fund Mythos Ventures, among others. In October, its R&D department Delphi Lab launched an AI accelerator in partnership with the NEAR Foundation, aimed at promoting the development of projects at the intersection of AI and Web3.
a16z crypto, a venture fund established by a16z for the crypto field, is often seen as a barometer of the industry. Its relevant leaders have publicly stated multiple times that they support the integration of crypto and AI projects. As early as March, media reports indicated that a16z was raising a new fund of up to $6 billion, focusing on AI investments, planning to allocate 15% of that fund specifically for AI infrastructure and application development. Among the 21 projects announced in the a16z Crypto fall startup accelerator (a16z CSX) in September, five projects were related to crypto AI, including Skyfire, the decentralized AI network OpenGradient, the GPU project Kuzco, the blockchain AI solution Banyan designed for enterprises, and PIN AI. The recently popular GOAT (truth_terminal) project and several AI Agents have received cryptocurrency funding support from a16z crypto partners.
As we entered the second half of the year, more crypto venture capital institutions announced their entry into the crypto AI field, with some institutions establishing dedicated AI funds and others increasing their investments in crypto AI projects.
On June 19, Pantera Capital announced it was raising a new $1 billion fund, with over $200 million allocated for AI-related projects. The fund's investment manager, Cosmo Jiang, stated that in the next 10 to 20 years, all crypto companies will become AI companies, and Pantera Capital has already invested 15% to 20% of its capital in AI-related blockchain projects in its early funds.
In August, Grayscale launched a new decentralized AI trust fund called 'Grayscale Decentralized AI Fund', focusing on investments in AI-related project tokens. The first batch of selected projects includes Bittensor (TAO), Filecoin (FIL), Livepeer (LPT), Near (NEAR), and Render (RNDR) among other AI project tokens, and the holdings will be adjusted each quarter based on portfolio performance. As of December 23, the Grayscale AI fund's holdings encompass Bittensor (TAO), Filecoin (FIL), Near (NEAR), Render (RNDR), and The Graph (GRT), managing a total asset value of approximately $1.5 million.
Also in August, Binance Labs' investment director Max Coniglio publicly stated its recent strategy: to promote the development of decentralized applications (DApps) driven by AI, and has continuously invested in the decentralized AI network Sahara AI and the AI robot platform MyShell. Max Coniglio believes that the combination of AI and blockchain is a force that cannot be underestimated. If the two can work together towards open-source protocols, they will jointly create a prosperous new era. In a recent interview, Binance co-founder He Yi expressed that the integration of AI and crypto is sparking a productivity revolution, with its impact far exceeding imagination. While blockchain technology primarily changes labor relations and benefit distribution models, AI is a revolutionary technology that directly enhances the productivity of users.
In October, Coinbase Ventures also publicly stated that its investment strategy has shifted from single cryptocurrency investments to in-depth layouts of 'Crypto + AI' projects. Its head Hoolie Tejwani pointed out that there is a natural complementarity between crypto technology and AI, with the two technologies being closely linked like the double helix structure of DNA, laying a solid foundation for the future digital life. In fact, as early as August, Coinbase Ventures had already participated in investments in AI Agent payment projects Skyfire and Payman.
The collective optimism of this series of crypto venture capital institutions towards the crypto AI field undoubtedly sends a strong signal: in future market cycles, venture capital institutions will launch large-scale and far-reaching layouts and bets in the crypto AI field.
According to the Q3 2024 crypto investment report released by Messari, crypto venture capital institutions injected over $213 million into AI projects during this quarter, marking a growth of 250% compared to the previous quarter and a year-on-year increase of 340%.
Furthermore, in the recently released Messari (2025 Crypto Outlook Report) on December 24, a noticeable trend emerged: investors' interest in the intersection of crypto and AI is growing stronger, with crypto AI becoming the most dazzling emerging investment theme of 2024. This field has not only achieved approximately 100% year-on-year growth in total financing but also witnessed a significant 138% increase in the number of financing rounds. Notably, well-known Web3 accelerator projects like a16z CSX and Beacon have included many projects in the crypto AI field on their selected lists. This data indicates that more and more venture capital investment firms are flooding into this emerging field, and the influx of capital will undoubtedly further accelerate the rise of the crypto AI track.
The AI investment strategy of crypto institutions: a preference for decentralized AI infrastructure and application products.
According to the 2024 investment data整理 from the Rootdata platform of crypto venture capital institutions, a16z, Delphi Venture, Coinbase Ventures, Pantera Capital, Binance Labs, Hack VC, Polychain, etc., all participated in multiple crypto AI project investments in 2024. Notably, Coinbase Ventures, Delphi Venture, Binance Labs, and Hack VC were particularly active, demonstrating strong optimism for this field through multiple investments.
Coinbase Venture has increased its investment in crypto AI projects since August, investing in the foundational infrastructures of two AI Agent payment projects, Payman and Skyfire, while Binance Labs and a16z Crypto have participated in multiple early projects through both direct investment and incubator programs.
As a venture capital and incubation arm of Binance, Binance Labs' projects are often seen by the crypto community as a wealth password for early listings. Its influence is profound. According to the disclosed crypto AI investment projects in 2024, its strategy is for projects to be selected for the MVB incubator before further investment, as seen with investments in NFPrompt (NFP) and MyShell, both of which emerged from the MVB incubator and are mainly AI application products.
a16z Crypto has a deep investment history in the AI space, having participated in over $100 million financing for several AI star companies like Open AI, AI chatbot Character.AI, and AI pharmaceutical company Genesis Therapeutics.
However, according to Rootdata's data, the number of crypto AI projects directly invested by a16z crypto is not many; only two projects are disclosed—AI Agent development network Axal and AI virtual human creation platform Balance.fun, with Axal being one of the projects funded through a16z CSX. It tends to participate more in the form of the a16z CSX incubator. Among the 21 projects announced in the a16z Crypto fall startup accelerator (a16z CSX) in September, five projects are related to crypto AI, including Skyfire, the decentralized AI network OpenGradient, the GPU project Kuzco, the blockchain AI solution Banyan designed for enterprises, and PIN AI. Several AI Agents, including the recently popular GOAT (truth_terminal) project, have received cryptocurrency funding support from a16z crypto partners.
In a recent forecast of 2025 trends, a16z emphasized that the intersection of crypto and artificial intelligence will be a key focus in the coming year, especially optimistic about the development of crypto AI Agents, believing that AI Agents will be able to own and manage their own wallets, operate on-chain transactions like node validation, and behave as true autonomous AI chatbots rather than being human-controlled, capable of achieving autonomous validation and other functions.
ChainCatcher's analysis of the project data involving investment institutions reveals that the projects participating in investment in 2024 are mostly AI technology infrastructure and application products, with institutions like Polychain and Pantera Capital tending to invest in AI infrastructure such as decentralized AI networks, GPU tokenization, and AI data, while Hack VC and Binance Labs prefer AI application products.
The list of crypto AI projects laid out by venture capital firms like Coinbase, a16z, and others in 2024.
Coinbase Venture's 2024 crypto AI project investment list
1. AI Agent payment tool Payman
Payman is an AI Agent payment proxy solution designed specifically for AI models and developers, created in May of this year, aimed at building a payment infrastructure for collaboration between AI and humans. Developers can integrate Payman's payment API interface to enable their AI Agents to autonomously handle payment transactions between AI and humans, whether it's AI developers wanting to pay for human assistance or humans wanting to be compensated for collaborating with AI. This can all be done through integrating Payman API, allowing AI Agents to automate the payment processes between both parties.
The core advantage of Payman's product is to provide AI developers with a complete and robust AI Agent payment proxy system that can autonomously and intelligently handle all payment processes, greatly simplifying operational complexity. Payman supports not only fiat currency payments but also the use of cryptocurrencies as payment methods, such as USDC.
On August 8 of this year, Payman announced the completion of $3 million in pre-seed financing, with investors including Visa, Coinbase Ventures, and The Spartan Group among other well-known venture capital institutions. Additionally, this project was selected for Binance Labs' MVB accelerator program as early as July.
Coinbase Ventures' preference for Payman is reflected not only in its financial support but also in its repeated public backing of the project. During the offline event of Devcon2024 hosted by Coinbase in November this year, Payman showcased how to use the USDC stablecoin for seamless payments between AI and humans in collaboration with virtual idols under South Korean entertainment giant HYBE.
It is also worth mentioning that the project's co-founder Tyllen Bicakcic previously held key developer relations positions at the Uniswap Foundation and Flow blockchain.
2. AI Agent payment system Skyfire: the Visa of the AI space.
Skyfire is also an AI Agent payment solution system that has received investment from Coinbase Ventures. This product was founded in June this year by a former executive of Ripple. It has designed a complete transaction process for AI Agents to operate independently, without any human intervention, such as purchasing and completing any goods required for specified tasks.
It is well known that AI Agents will be able to perform functions such as autonomous shopping, booking tickets, route planning, and app listing. However, typical AI Agents cannot autonomously complete the final payment process, such as bills, tickets, website hosting fees, and other expenses. Skyfire was designed to solve this pain point, building a comprehensive cryptocurrency payment system for AI Agents, enabling them to autonomously conduct financial operations such as payments and receipts without any human intervention. For example, when a user provides funds to an AI Agent, it can autonomously use those funds to meet user needs, with the entire process requiring no human involvement, just like a human assistant using cash or credit cards to buy coffee. Therefore, Skyfire is regarded as the 'Visa' payment center in the AI field, supporting AI Agents in autonomously handling payments, receipts, and savings.
In terms of financing, Skyfire also performed well, receiving strong support from several well-known crypto capitals. In August of this year, it successfully completed $8.5 million in seed round financing, attracting investments from well-known institutions like Circle, Ripple, and Gemini. In September, Skyfire was selected for a16z's fall CSX accelerator. In October, Coinbase Ventures and a16z Crypto CSX announced a $1 million new strategic investment in Skyfire.
As of December 21, the publicly accumulated financing amount for Skyfire has reached $9.5 million.
3. Decentralized AI data network Vana
Vana is a decentralized data network aimed at building a platform allowing users to tokenize and trade their personal data and AI models. The project originated from a research project at MIT in 2018, dedicated to constructing a user-owned data pool network, enabling users not only to truly own and control their data but also to autonomously build AI models, freely deciding how to use their data and earning economic returns through data contributions.
Vana has built a decentralized data market for users, allowing them to control their social data autonomously through smart contracts. Whenever their data is used, users automatically receive rewards. For instance, users can provide high-quality data generated on platforms like Google and Reddit to AI developers for model training and earn income from it. This not only effectively addresses the problem of scarcity of training data for AI models but also opens a new way for users to earn income from their own data.
On September 18, Vana announced the completion of $5 million in strategic financing, led by Coinbase Ventures. In April of this year, Vana revealed that it had previously secured $18 million in Series A financing led by Paradigm in October 2022, which had never been disclosed before, and stated that its total financing amount had reached $20 million.
As of December 21, Vana's total financing amount reached $25 million.
On December 13, Vana (VANA) was launched on Binance Launchpool.
4. Decentralized AI network OpenGradient
OpenGradient aims to create a decentralized AI network that supports developers to create, distribute, and deploy AI models and applications in an open and permissionless manner. OpenGradient not only provides developers with AI development tools but also facilitates the secure and seamless integration of AI workflows into web3 applications, and enables decentralized and generalized access to AI models.
Currently, OpenGradient is building a blockchain network compatible with EVM, allowing developers to deploy AI models using Solidity smart contracts. It is seen as the 'HuggingFace' of the Web3 space. AI developers' GitHub, Hugging Face is an open-source model repository that supports users in sharing various machine learning models, referred to as 'the GitHub of the AI world.'
In September of this year, OpenGradient was one of the 21 selected companies in the a16z Crypto fall startup accelerator (CSX).
On October 10, OpenGradient announced the completion of $8.5 million in seed round financing, with investors including a16z CSX, Foresight Ventures, Coinbase Ventures, Symbolic Capital, and others participating.
5. Decentralized AI model production factory Pond
Pond is building a decentralized AI model network that allows any user to easily create their own AI models and tokenize them for profit, regarded as a 'decentralized AI model production factory.'
In the Web2 world, there is a data modeling and analysis competition platform called Kaggle, which supports developers to compete in building the best AI models. However, creators do not have final ownership rights. Pond aims to address this issue by providing developers with a range of foundational infrastructure needed to create AI models, allowing creators to own, control, and tokenize their AI models, and is seen as a 'Web3 improved version of Kaggle.'
Unlike most projects based on LLM models, Pond is based on a large-scale GNN graph neural network to perform real-time statistics and predictions on on-chain data. Compared to LLM, it excels at processing and generating information. GNN-based Pond is better at mining relationships between data and extracting valuable information, leveraging on-chain data native to the crypto industry, and using technologies such as graph neural networks to learn and predict on-chain behavior. Many new business logics have emerged from predicting behavior, such as token price prediction, AI-enhanced MEV, and DeFi strategies.
On November 7, Pond announced the completion of $7.5 million in seed round financing, with investment institutions including Archetype, Delphi Ventures, and Coinbase Ventures.
a16z Crypto and a16z CSX accelerator's 2024 crypto AI project investment list:
1. AI Agent development network Axal
Axal is dedicated to building a verifiable AI Agent development network for any task, providing users with a complete set of AI Agent creation tools, allowing users to create their own AI Agent intelligent systems by stating their intentions.
In October, Axal announced it had completed $2.5 million in pre-seed financing, led by CMT Digital, with other participants including a16z Crypto Startup School, Escape Velocity, IDG Vietnam, and Artichoke Capital.
Meanwhile, Axal launched its flagship product, Axal Autopilot, an AI Agent-driven trading automation platform that provides users with personalized trading strategies, including price tracking, on-chain trade execution, and yield management.
2. AI virtual human creation platform Balance.fun (EPAL)
Balance was originally launched in February this year by the E-PAL gaming companion platform as a Web3 experience infrastructure, aiming to provide a smooth transition for Web2 users to Web3. The Balance platform incorporates an AI-driven system as a core element, allowing users to create personalized AI virtual companions that can participate in gaming, learning, entertainment, and even productivity tasks, adapting to personal preferences, behaviors, and needs.
In September of this year, E-PAL disclosed that it has successfully completed two rounds of financing led by a16z and Galaxy Interactive, with a total financing amount of $30 million. In November, Balance.fun announced that it had raised over $10 million in sales from institutions through early node sales, with participating institutions including Animoca Brands, Amber Group, GSR Markets, and others.
Currently, Balance.fun is selling nodes at a price of $599 each. According to the roadmap, it is expected to issue the token ETP in January 2025.
3. Decentralized personal AI network infrastructure PIN.AI
PIN AI aims to create an open AI network that can access vast contextual data without restrictions, enabling AI builders to create various useful personal AI applications. It is well known that the inference of AI models is limited by the input data; however, due to privacy concerns, much data resides on users' personal devices such as phones and computers. PIN AI aims to mobilize this personal data and support users in creating their own personal AI applications based on their needs, allowing AI developers to provide very practical AI services for everyday tasks like shopping, organizing travel, and financial planning.
The main goal of the PIN protocol is to achieve an open ecosystem for personal AI applications. It comprises three foundational layers: personal data (focusing on privacy and data ownership), personal AI (a useful and trustworthy companion AI on personal devices), and external AI (an open market for AI services).
This allows AI applications to easily access user contextual data and match user intentions with specialized external AIs while always protecting user privacy.
On September 10, PIN AI announced the completion of $10 million in pre-seed financing, with participation from a16z CSX, Hack VC, Blockchain Builders Fund (Stanford Blockchain Accelerator), and others.
Binance Labs' 2024 crypto AI project investment list:
1. AI content creation platform NFPrompt (NFP)
NFPrompt (NFP) is an AI-driven NFT content generation platform that provides creators with AI content creation work such as images, videos, pfp, music, etc. Users can easily create stunning works through prompts and generate NFTs for ownership confirmation with one click. It was selected for Binance Labs' sixth incubation program in September 2023 and launched its token NFP on Binance in December of last year, with investment from Binance Labs announced in March this year.
2. Decentralized AI data network Privasea
Privasea, formerly known as Nulink, primarily focused on privacy data computation. It was selected for Binance Labs' fourth incubation program as early as May 2022 and now provides solutions for the privacy issues of AI computing data, supporting data privacy and security during AI model computation. Additionally, its identity verification program ImHuman functions similarly to WorldCoin and supports converting collected human biometric features into AI model training data in a privacy-secure manner.
In March of this year, Privasea announced it had completed $5 million in seed round financing, with participation from Binance Labs and others.
3. AI knowledge sharing and search platform QnA3.AI (GPT)
QnA3.AI (GPT), abbreviated as QnA3, is an AI-driven Web3 knowledge sharing and search platform aimed at providing users with accurate knowledge in the cryptocurrency field. It not only offers information access and retrieval but also provides complex logical analysis functions to assist users in decision-making for trading and more. It was also one of the projects selected for Binance Labs' sixth incubation program in September 2023, and the QnA3.AI token GPT was launched on OKX in February, with Binance Labs' investment announced in March.
4. AI robot platform MyShell
MyShell is building an AI consumer application layer that connects users, creators, and open-source AI researchers, providing a complete set of AI Agent creation tools that allow users to easily discover or create various applications or robots based on AI Agent technology, such as voice chatbots, language learning assistants, image generation tools, etc. It is also one of the selected projects in Binance Labs' sixth incubation program for 2023, and completed a $5.6 million seed round financing at a valuation of $57 million in December of the same year.
In March of this year, MyShell announced the completion of $11 million in Pre-A financing, with investors including Dragonfly, Delphi Digital, Bankless Ventures, and others. In August, it announced further investment from Binance Labs.
5. Decentralized AI network Sahara AI
Sahara AI is a decentralized AI network infrastructure that facilitates AI assetization, aimed at helping users deploy or build customized and personalized AI products, and ensures that all contributors in the AI chain receive fair attribution and rewards.
In August of this year, Sahara AI announced the completion of $43 million in financing, with institutions such as Binance Labs, Pantera Capital, and Polychain Capital leading the investment.
In the roadmap published on December 24, Sahara AI stated that it has launched a data service platform and testnet, allowing users to earn rewards through data collection and annotation. The Sahara Chain mainnet is expected to launch in Q3 2025.
Pantera Capital's 2024 crypto AI project investment list: 1. Decentralized AGI system Sentient
Sentient is committed to building a decentralized artificial general intelligence (AGI) platform through community contributions, aiming to compete with traditional closed-source AI development models like OpenAI. It supports AI model creators in tokenizing their models and incentivizes users to contribute and expand AI models using Web3 technology, jointly building open-source AI models contributed by the community.
In July of this year, Sentient announced the completion of $85 million in seed round financing, with investors including Pantera Capital, Framework Ventures, Delphi Ventures, Hack VC, and others.
Delphi Digital's 2024 crypto AI project investment list:
1. Decentralized AI operating system 0G Labs
OG Labs (full name Zero Gravity Labs) initially appeared with a modular AI blockchain narrative, stating that it has a scalable DA layer to support AI DApps. In March of this year, OG announced the completion of $35 million in Pre-Seed financing, with investors including Hack VC, Animoca Brands, and Delphi Digital.
However, in November, OG Labs announced that it had obtained $290 million in new funding, which includes $40 million in seed round financing and a $250 million token purchase commitment. OG Labs co-founder Michael Heinrich stated that once the OG token is listed and circulated on cryptocurrency exchanges next year, the project will be able to extract funds from the token commitments.
Meanwhile, OG Labs has transitioned from a modular AI blockchain to building what it calls a decentralized AI operating system (dAIOS), aimed at enabling the development of on-chain AI applications, with the previously mentioned modular AI being just a part of it.
According to co-founders' public statements, the mainnet launch of the 0G protocol is expected in the first or second quarter of next year.
2. AI data collection platform Grass (GRASS)
Grass is a decentralized network resource sharing and redistribution DePIN platform developed by Wynd Network (a decentralized internet proxy network), allowing people to sell idle internet bandwidth and network data.
Users can automatically share idle bandwidth and personal data by downloading the Grass Chrome browser plugin and receive additional rewards, similar to a new concept of earning tokens by utilizing traffic and data. Additionally, Grass will convert the collected personal data into AI datasets for developers to use for AI training. Therefore, Grass is also seen as a decentralized AI data provisioning layer.
In September of this year, Grass announced the completion of Series A financing, reportedly valued at nearly one billion dollars, with investors including Hack VC, Polychain Capital, Delphi Digital, and others, although the specific financing amount has not been disclosed. On October 16, the Grass token GRASS was officially launched.
On November 25, Mechanism Capital partner Andrew Kang stated that the development of AI has made data one of the most valuable commodities in the world. Grass has created a data pipeline that most AI companies cannot reach through a crypto incentive mechanism. At the current growth rate, it is expected to achieve nine-digit revenue levels next year, possibly even exceeding billions of dollars.
3. Decentralized AI network Allora
Allora, formerly known as Upshot, was originally an NFT stock price protocol, renamed to Allora in February this year, transitioning from an NFT evaluation platform to a decentralized AI network.
On the Allora platform, users can evaluate the performance of an AI model under current conditions by using multiple AI Agents to predict tasks and assign rewards based on the quality of agent predictions and evaluations. This incentive mechanism allows models on Allora to continuously learn and improve, adjusting as the market evolves.
In June, Allora announced the completion of $3 million in strategic financing for the decentralized artificial intelligence network Allora Labs, with investors including Archetype, Delphi Ventures, CMS Holdings, etc. Thus, the total financing of the Allora project has reached $35 million.
4. Decentralized GPU project io.net (IO)
io.net is a decentralized AI computing DePIN protocol aimed at addressing the insufficient computing demands in AI and machine learning by utilizing distributed GPU and CPU resources. It has built a bilateral market around chips, with the supply side comprising globally distributed GPU and CPU computing power, and the demand side consisting of AI engineers seeking to complete tasks such as AI model training or inference. Its mission is to integrate millions of GPUs into its DePIN network.
In March, io.net announced the completion of $30 million in Series A financing, with investors including Hack VC, Multicoin Capital, Delphi Digital, Solana Labs, etc. Thus, the project's total financing has reached $40 million, with a valuation of $1 billion.
In June, the IO token of io.net was issued on Binance Launchpool; in October, the IO token was launched on Coinbase.
Hack VC's 2024 crypto AI project investment list:
1. GPU tokenization platform Exabits
Exabits aims to unlock over $1 trillion in AI computing market value by tokenizing physical GPUs into liquid financial assets that can be invested. By splitting and tokenizing GPU computing resources, investors can purchase EGPUs representing GPU capacity as assets to earn profits.
Currently, the Exabits platform has created a pathway to GPU-Fi (G-Fi), allowing GPU assets to flow between various financial platforms, enhancing the capital efficiency of the entire AI ecosystem while achieving the democratization of AI computing, enabling anyone to participate in the AI computing economy. According to official data, Exabits has provided enterprise-level GPU resources for companies such as EMC, Viggle, Lepton, IO.NET, Aethir, Akash, MyShell, Nebula Block, and Game Killer.
In February of this year, Exabits announced the completion of seed round financing, with specific amounts not disclosed. On December 11, Exabits announced that it had completed $15 million in a $15 million seed round financing, led by Hack VC. This financing brought Exabits' total financing to $20 million.
2. GPU asset tokenization platform GAIB
GAIB, like Exabits, aims to unlock a trillion-dollar AI computing market by tokenizing GPU assets, changing the traditional GPU financing model. On December 13, it announced the completion of $5 million in seed round financing, with participating institutions including Hack VC, Faction, Aethir, and others.
According to official disclosures, GAIB has completed prototype development and plans to launch a pilot project this month, with plans to release the Alpha product and platform token within the next two months.
3. AI Agent infrastructure Theoriq
Theoriq is a modular, composable AI Agent infrastructure network launched by the Web3 machine learning platform ChainML, supporting any user to create their own AI Agent without coding through the provision of relevant AI Agent development infrastructure tools.
In May, the developer of Theoriq, ChainML, announced it had raised $6.2 million in seed round expansion financing, which was led by Hack VC.
Currently, Theoriq is in testnet status, and according to the roadmap, it is expected to go back online on the mainnet by the end of 2024.
4. AI image generation robot imgnAI
imgnAI is a consumer-facing AI image generation robot, allowing users to automatically generate images through simple text commands, similar to text-to-image models like Midjourney. However, users on the imgnAI platform need to purchase the native token IMGNAI to unlock more creative tools and functions.
In January of this year, imgnAI announced the completion of $1.6 million in seed round financing, led by Hack VC, and also developed the Twitter AI Agent virtual persona @Naifu.
Polychain's 2024 crypto AI project investment list:
1. Decentralized AI cloud platform Hyperbolic
Hyperbolic is building an open AI cloud platform, planning to integrate global GPU computing resources through blockchain technology to address the industry's bottleneck in AI model computation costs and distribution. To this end, Hyperbolic has built a specialized AI inference service platform and a GPU marketplace, the former allowing developers to easily call open-source AI models through simple API calls while supporting AI model training and fine-tuning, and the latter enabling GPU computing power to be traded like goods, supporting users to rent or buy GPU.
On December 11, Hyperbolic announced the completion of $12 million in Series A financing, led by Variant and Polychain Capital. Earlier in July, Hyperbolic announced it had completed $7 million in seed round financing, also led by Polychain Capital. Thus, Hyperbolic's total financing has reached $20 million.
2. L1 network Talus Network designed for AI Agents
Talus Network is an L1 network infrastructure designed specifically for AI Agents, enabling any user to create their own dedicated AI Agent. It is important to note that this L1 network is designed based on the Move language and is not yet live on the mainnet, but applications for joining the testnet are available.
On November 26, Talus Network announced the completion of $150 million in financing with $6 million raised, led by Polychain Capital. Earlier in February this year, it also completed $3 million in financing, again led by Polychain Capital. As of December 24, Talus Network's publicly disclosed total financing has reached $9 million.
3. Decentralized AI data network OpenLedger
OpenLedger is a blockchain for AI data, providing the infrastructure for creating specialized language models for specific domains, supporting developers in creating various applications such as AI Agents, chatbots, and smart driving. It collects and manages specific data through its data network, using blockchain technology to attribute reward model data to contributors and creators.
On July 2, OpenLedger announced the completion of $8 million in seed financing, led by Polychain Capital.
4. Decentralized AI network Ritual
Ritual is a decentralized AI network infrastructure aimed at creating a system that allows for the open and permissionless creation, training, improvement, and trading of AI models. Its provided SDK not only supports developers in seamlessly integrating AI models into applications or protocols on any chain, but also allows users within the network to fine-tune AI models, tokenize trades, and conduct inference training using cryptographic schemes.
On April 8, Ritual announced it secured several million dollars in Series A1 financing from Polychain Capital, with the specific investment amount undisclosed. As early as November 2023, it completed $25 million in financing from Archetype and others.
The Ritual Chain testnet went live on November 19, providing developers with a platform to build AI-native applications.
