Let’s Get Agentic: The State of Crypto AI Agents in 2025

Kevin Dwyer
February 18, 2025
9 min read
Imagine using software that doesn’t just follow your commands but thinks, learns, and adapts—advancing your goals autonomously. This is exactly what AI agents are designed to do, whether that’s optimizing trading, scheduling appointments, prompting actions on other platforms, or a wide variety of other tasks.
Crypto AI agents take this idea a step further by acting autonomously in the world of blockchain. They are quickly transforming trading, DeFi, and the whole arena of Web3, performing read-and-write interactions on-chain and turning decentralized decision-making into a self-evolving ecosystem. As AI and blockchain intersect, they are reinventing how we create, exchange, and secure digital value.
So let’s take a closer look, and let’s get agentic!
But first, a word of motivation for potential Crypto AI agent builders:
“It’s so much easier to be an agent developer (or any kind of developer) than it ever was. You should start now. You should change your life. You can do this. This is a time of radical change and if you start now, you are early at ground zero.”
– Shaw, Founder of Eliza Labs on Bankless
What Exactly Are Crypto/Web3 AI Agents?
An AI agent is an autonomous system that perceives its environment, processes information, and uses artificial intelligence models to take actions to achieve specific goals (usually on behalf of a user).
Taking that concept to the next level, crypto AI agents are specifically designed to automate and execute tasks on-chain, such as high-frequency trading and all other forms of blockchain interactions.
Some more definitions broken down:
AI Agent → A smart software program that can think and act on its own.
Powered by large language models (LLMs) as a cognitive core → It uses advanced AI (like ChatGPT or DeepSeek) to understand and process information, acting as its "brain."
Intelligent ‘agentic’ system → This means crypto AI agents are designed to make decisions and take actions by themselves.
Capable of autonomously controlling and determining the execution paths → The AI can decide what next steps to take without needing constant human input.
Armed with blockchain execution capabilities → They can retrieve data from a blockchain and also call smart contract functions to execute transactions that are recorded to the blockchain.
For example:
- The agent analyzes market data and decides to make a trade.
- It generates a transaction (e.g., swapping SOL for USDC).
- It calls a smart contract function (e.g., swapExactTokensForTokens() on Uniswap).
- The transaction is signed and broadcasted to the blockchain.
Under user’s instructions → Agents follow general goals or tasks set by a person but figure out the best way to complete them on their own.
How Do You Use Crypto AI Agents?
There are a variety of ways you can interact with and use Crypto AI agents, including Chat interfaces like Telegram, Slack, Discord.
Interacting with crypto AI bots on platforms like Telegram and Discord is simple and typically involves commands, buttons, and natural language queries. These bots help automate trading, provide real-time market data, and manage your crypto portfolio—all from within your chat interface.
Example Commands:
/balance → Check your wallet balance.
/trade BTC 0.5 ETH → Trade 0.5 BTC for ETH.
/portfolio → View your AI-managed crypto portfolio.
/price ETH → Get the current ETH price.
/auto_trade on → Activate AI-driven auto-trading.
/market_trends → Fetch AI-generated market insights.
/gas_fees → Show current Ethereum gas fees.
Web Chatbots & AI Assistants
Many crypto AI agents have web-based chat interfaces where you can type commands or ask AI questions directly.
Example Inputs:
- "Monitor and alert for whale transactions above $10M on-chain."
- "Set up auto-buy of 0.1 BTC every month."
- "Backtest a trading strategy for Solana over the past 12 months."
Some platforms even offer voice-based AI assistants for hands-free commands!
API-Based Chat Interfaces (Advanced Users & Developers)
If you’re a developer, you can integrate crypto AI agents via APIs and interact with them through a CLI (command-line interface) or custom applications.
Example API Calls:
json
{
"action": "trade",
"pair": "ETH/USDT",
"amount": "1.5",
"strategy": "AI_predicted"
}
- Example CLI Commands:
- $ ai-agent trade BTC 1 ETH
- $ ai-agent predict BTC
- $ ai-agent stake 500 DAI
Specific Benefits of Crypto AI Agents to Users
Crypto AI agents can be employed to complete a huge variety of tasks.
- Automated Trading & Portfolio Management – Execute trades at optimal times 24/7 using AI strategies, reducing human error and emotional decision-making.
- Real-Time Market Analysis – Process vast amounts of blockchain and off-chain data instantly to provide insights and predictions.
- DeFi Yield Optimization – Automatically find and allocate funds to the most profitable DeFi lending, staking, or liquidity pools.
- Fraud Detection & Security – Identify suspicious transactions, prevent scams, and enhance on-chain security.
- Smart Contract Automation – Execute complex smart contract functions without manual intervention, improving efficiency and reliability.
- Lower Transaction Costs – Optimize gas fees and transaction execution to save costs.
- Personalized Financial Strategies – Tailor investment and risk management strategies based on user preferences and market conditions.
- Cross-Chain Interoperability – Bridge assets and data across different blockchains, enabling seamless decentralized finance interactions.
- Reduced Cognitive Load – Simplifies decision-making for users by analyzing trends, filtering noise, and executing predefined strategies.
Your Directory To the Crypto AI Agent Universe
If you think this is all hypothetical, think again. Some of the smartest teams in Web3 are already building AI-driven automation:
DeFi & Trading Agents
DeFi AI agents automate decentralized finance (DeFi) tasks like yield farming, lending, and portfolio management while trading AI agents execute trades, analyze market trends, and optimize strategies based on real-time data—both helping users maximize returns and minimize risks with minimal manual effort.
- Polytrader: AI assistant for Polymarket, analyzing sentiment for better prediction market decisions.
- Zotto: An Intent-Driven Smart Agent Hub with a built-in Autonomous PerpDEX and powered by AI.
- Axal: Builds a verifiable autonomous agent network for secure decentralized interactions.
- Almanak: Enables AI-driven DeFi strategies by optimizing autonomous financial agents.
- Wayfinder: Connects AI with blockchains for seamless cross-chain interactions.
- The Hive: Modular DeFi agent network automating complex on-chain operations.
- Hey Anon: AI-driven DeFi tool simplifying data aggregation and execution.
Meme & Influencer Agents
Meme and influencer AI agents, like Truth Terminal and AIXBT, generate viral content, analyze trends, and engage with audiences by automating meme creation, market commentary, and social media interactions—shaping narratives, driving engagement, and influencing crypto communities.
- Terminal of Truths: One of the first meme-inducing AI agents to generate content and engage users on social media.
- Aixbt: AI-driven crypto market intelligence and narrative analysis.
- Dolos: An AI influencer and autonomous X poster built on Dolion AI.
- Thales: Data-driven analyst & autonomous onchain trader built on ElizaOS.
Investment DAOs
AI investment DAOs like the following leverage artificial intelligence and decentralized governance to fund emerging projects. These DAOs use AI-driven decision-making and community governance to allocate capital efficiently in the Web3 economy.
- AIMonica: AI-driven DAO investing in meme assets and AI projects using decentralized governance.
- Vader: AI agent-led DAO investing in small-cap Web3 projects with decentralized decision-making.
- aixCB: AI-powered VC DAO funding AI and DeSci startups with automated investment insights.
- Sekoia: DAO fund investing in AI and blockchain projects through decentralized governance.
Generative AI Agents
Generative crypto AI agents create AI-driven art, music, and digital content on the blockchain. They generate unique on-chain assets, NFTs, and interactive media while using smart contracts for ownership and provenance, redefining creativity in Web3.
- Zerebro: AI agent creating and distributing content in music, NFTs, and social media.
- Luna: AI agent and “Virtual AI-Dol” singer with an EP on Spotify built with the Virtuals G.A.M.E agent framework.
- Sandy: An AI Video Agent that brings brands to life with human-like video interactions.
- Nft_xbt: AI on-chain detective sharing crypto insights and alpha.
Frameworks for Creating & Launching New Agents
- elizaOS: Open-source framework for building and managing AI agents in TypeScript.
- Codex: A platform to build no-code AI agents to integrate with dApps, sites, and games.
- Zerepy: Python framework for deploying AI agents in NFTs, music, and social platforms.
- ARC Rig: Rust-based framework for high-performance AI agents, optimized for Solana.
- G.A.M.E: AI agent framework with APIs and SDKs for structured decision-making.
- Virtuals: Decentralized platform to create, tokenize, and co-own AI agents.
Data Providers for AI Agents
- Ankr: Offers high-performance blockchain APIs enabling AI agents to access real-time on-chain data.
- Vana: Provides a decentralized data platform where contributors can share data for AI model training.
- Sahara AI: Open data marketplace with high-quality datasets to enhance AI accuracy.
- Masa: A decentralized data network allowing contributors to earn by providing data for AI applications.
How Ankr is Powering the Agentic Wave
Providing Templates For Building New Crypto AI Agents
Ankr's blockchain APIs, integrated within the elizaOS framework, empower developers to rapidly prototype and deploy AI agents capable of executing an array of blockchain and crypto-related tasks. By plugging Ankr’s pre-built API connections into elizaOS's modular library, developers gain immediate access to blockchain data, including token balances, gas price predictors, NFT data, and all other critical data. This seamless integration eliminates the need for extensive custom coding, drastically accelerating the agent development process. Ankr’s Advanced API also provides access to indexed historical chain data, enabling developers to train their AI agents on real-world market dynamics like MEV patterns and liquidity trends.
Learn more about our work with elizaOS
Giving AI Agents A Stream of Data From 70+ Blockchains
Ankr’s Web3 API and RPC connections empower crypto AI agents with instant blockchain access for seamless read-and-write operations across multiple chains. The Advanced API provides indexed historical data, enabling AI to analyze MEV patterns, liquidity trends, and market dynamics. By integrating this API, AI agents can query multiple networks in one request, access pre-indexed data for near-instant results, and leverage globally distributed bare-metal servers for ultra-low latency. These capabilities enhance AI-driven crypto applications, optimizing decision-making and execution for maximum efficiency in decentralized markets.
Leading the Way In Creating In-House Agents
“Every business is going to have an AI agent that their customers can talk to in the future.”
– Zuck
Ankr has already created our own Web3 AI Agent called Moniqa to assist in an array of tasks from both the internal team and our clients. Moniqa specializes in retrieving and analyzing real-time and historical telemetry data for Ankr’s blockchain infrastructure. This includes:
- Detailed RPC usage metrics, like request rates, latency, and error distribution.
- Insights into geographic traffic patterns, helping us pinpoint usage trends and optimize node placement.
- Granular request logs that reveal call details, tenant/API key activities, and node performance.
- Comprehensive health data on blockchain nodes, including sync statuses and resource usage.
This data is invaluable for the Ankr team, as it helps them troubleshoot issues, plan capacity, and ensure our nodes run smoothly. For external clients, it demonstrates our commitment to transparency and robust performance management.
Where Are Agents Headed From Here?
If Web3 is about decentralization and AI is about automation, then crypto AI agents are the inevitable intersection of both. But let’s be real: we’re still in the early days. There’s a lot to figure out—security concerns, regulatory challenges, and the small matter of making sure AI agents don’t go rogue and wreck entire markets.
That said, the opportunity is massive. Web3 AI agents are already proving their value, and as infrastructure catches up, we’re going to see an explosion of use cases that go way beyond trading and memetics. So whether you’re a developer, investor, or just someone trying to keep up with crypto’s endless innovations—pay attention. The agents are here. And they’re smarter than you think!
Join the Conversation on Our Channels!
Twitter | Telegram | Substack | Discord | YouTube | LinkedIn | Reddit | All Links
Similar articles.

How AI and Blockchain Work Together: Use Cases for Complementary Technologies

Kevin Dwyer
October 31, 2023
Early this year, the hype around artificial intelligence became an unstoppable force as mysterious new projects like Open AI made headline after headline. The immediate...

How We’re Making Blockchain-Aware AI Agents With elizaOS

Kevin Dwyer
February 4, 2025
TL;DR
- elizaOS is a powerful framework for building AI agents.
- Ankr lets agents learn from blockchain data and execute on-chain tasks autonomously.
- **The possibilities are...