标题：DeFai：人工智能与 DeFi 的融合

短短 3 个月内，AI x memecoin 的市值已达到 134 亿美元，其规模与 AVAX 或 SUI 等一些成熟的 L1 相当。

output: Artificial intelligence (AI) has experienced a surge in popularity within the cryptocurrency industry, particularly on the Solana blockchain, where the integration of AI and memecoins has led to a market capitalization of $13.4 billion, comparable to some mature Layer 1 protocols such as Avalanche (AVAX) or Sui (SUI).

产出：人工智能 (AI) 在加密货币行业中的受欢迎程度激增，特别是在 Solana 区块链上，人工智能和 memecoin 的整合导致市值达到 134 亿美元，与一些成熟的 Layer 1 协议（例如雪崩 (AVAX) 或隋 (SUI)。

Over the past few months, we've witnessed the merging of two powerful technologies: AI and blockchain. From decentralized model training on the early Bittensor subnet to decentralized GPU/computing resource markets like Akash and io.net and the current wave of AI x memecoins and frameworks on Solana, each stage demonstrates how cryptocurrency can complement AI to a certain extent through resource aggregation, thereby achieving sovereign AI and consumer use cases.

在过去的几个月里，我们见证了两种强大技术的融合：人工智能和区块链。从早期Bittensor子网上的去中心化模型训练，到Akash、io.net这样的去中心化GPU/计算资源市场，再到当前Solana上的AI x memecoins和框架浪潮，每个阶段都展示了加密货币如何通过资源聚合在一定程度上补充AI ，从而实现主权人工智能和消费者用例。

In the first wave of Solana AI coins, some have brought meaningful utility beyond pure speculation. We’ve seen the emergence of frameworks like ai16z’s ELIZA, AI agents like aixbt that provide market analysis and content creation, or toolkits that integrate AI with blockchain capabilities.

在第一波 Solana AI 代币中，有些代币带来了超越纯粹投机的有意义的效用。我们已经看到了像 ai16z 的 ELIZA 这样的框架，像 aixbt 这样提供市场分析和内容创建的人工智能代理，或者将人工智能与区块链功能集成的工具包的出现。

In the second wave of AI, as more tools mature, applications have become the key value driver, and DeFi has become the perfect testing ground for these innovations. To simplify the expression, in this study, we refer to the combination of AI and DeFi as "DeFai".

在人工智能的第二波浪潮中，随着更多工具的成熟，应用程序已成为关键的价值驱动因素，而 DeFi 已成为这些创新的完美试验场。为了简化表达，在本研究中，我们将 AI 与 DeFi 的结合称为“DeFai”。

According to Coingecko, DeFai has a market cap of about $1 billion. Griffian dominates the market with a 45% share, while $ANON accounts for 22%. This track began to experience rapid growth after December 25, and frameworks and platforms such as Virtual and ai16z experienced strong growth after the Christmas holiday.

据 Coingecko 称，DeFai 的市值约为 10 亿美元。 Griffian 以 45% 的份额占据市场主导地位，而 $ANON 则占 22%。该赛道在12月25日之后开始快速增长，Virtual、ai16z等框架和平台在圣诞假期后出现强劲增长。

This is just the first step, and the potential of DeFai goes far beyond this. Although DeFai is still in the proof-of-concept stage, we cannot underestimate its potential. It will use the intelligence and efficiency that AI can provide to transform the DeFi industry into a more user-friendly, intelligent and efficient financial ecosystem.

这只是第一步，德辉的潜力远不止于此。尽管DeFai仍处于概念验证阶段，但我们不能低估它的潜力。它将利用人工智能所能提供的智能和效率，将 DeFi 行业转变为更加人性化、智能和高效的金融生态系统。

Before we dive into the world of DeFai, we need to understand how agents actually work in DeFi/blockchain.

在深入了解 DeFai 的世界之前，我们需要了解代理在 DeFi/区块链中的实际工作方式。

Artificial Intelligence Agent (AI Agent) refers to a program that can perform tasks on behalf of users according to workflow. The core behind AI Agent is LLM (Large Language Model), which can respond based on its training or learned knowledge, but this response is often limited.

人工智能代理（AI Agent）是指能够按照工作流程代表用户执行任务的程序。 AI Agent背后的核心是LLM（大型语言模型），它可以根据其训练或学到的知识做出响应，但这种响应往往是有限的。

Agents are fundamentally different from robots. Robots are usually task-specific, require human supervision, and need to operate under predefined rules and conditions. In contrast, agents are more dynamic and adaptive, and can learn autonomously to achieve specific goals.

代理与机器人有着根本的不同。机器人通常是特定于任务的，需要人工监督，并且需要在预定义的规则和条件下运行。相比之下，智能体更具动态性和适应性，并且可以自主学习以实现特定目标。

To create more personalized experiences and more comprehensive responses, agents can store past interactions in memory, allowing the agent to learn from the user’s behavioral patterns and adjust its responses, generating tailored recommendations and strategies based on historical context.

为了创建更个性化的体验和更全面的响应，代理可以将过去的交互存储在内存中，从而使代理能够从用户的行为模式中学习并调整其响应，从而根据历史背景生成量身定制的建议和策略。

In blockchain, agents can interact with smart contracts and accounts to handle complex tasks without constant human intervention. For example, in simplifying the DeFi user experience, including one-click execution of multi-step bridging and farming, optimizing farming strategies for higher returns, executing transactions (buy/sell) and conducting market analysis, all of these steps are completed autonomously.

在区块链中，代理可以与智能合约和账户交互来处理复杂的任务，而无需持续的人工干预。例如，在简化 DeFi 用户体验方面，包括一键执行多步骤桥接和挖矿、优化挖矿策略以获得更高回报、执行交易（买入/卖出）和进行市场分析，所有这些步骤都是自主完成的。

According to @3sigma’s research, most models follow 6 specific workflows:

根据 @3sigma 的研究，大多数模型都遵循 6 个特定的工作流程：

1. Data Collection

1. 数据收集

First, models need to understand the environment in which they work. Therefore, they need multiple data streams to keep the model in sync with market conditions. This includes: 1) On-chain data from indexers and oracles 2) Off-chain data from price platforms, such as data APIs from CMC/Coingecko/other data providers.

首先，模型需要了解它们工作的环境。因此，他们需要多个数据流来使模型与市场状况保持同步。这包括： 1) 来自索引器和预言机的链上数据 2) 来自价格平台的链下数据，例如来自 CMC/Coingecko/其他数据提供商的数据 API。

2. Model Reasoning

2. 模型推理

Once models have learned the environment, they need to apply this knowledge to make predictions or actions based on new, unrecognized input from the user. Models used by agents include:

一旦模型了解了环境，它们就需要应用这些知识来根据用户新的、未识别的输入做出预测或采取行动。代理使用的模型包括：

3. Decision Making

3. 决策

With trained models and data, the agent can take action using its decision-making capabilities. This includes interpreting the current situation and responding appropriately.

借助训练有素的模型和数据，代理可以利用其决策能力采取行动。这包括解释当前情况并做出适当反应。

At this stage, the optimization engine plays an important role in finding the best results. For example, before executing a profit strategy, the agent needs to balance multiple factors such as slippage, spread, transaction costs, and potential profits.

在这个阶段，优化引擎在寻找最佳结果方面发挥着重要作用。例如，在执行盈利策略之前，代理需要平衡滑点、点差、交易成本和潜在利润等多种因素。

Since a single agent may not be able to optimize decisions in different domains, a multi-agent system can be deployed to coordinate.

由于单个智能体可能无法优化不同领域的决策，因此可以部署多智能体系统进行协调。

4. Hosting and operation

4. 托管及运营

Due to the computationally intensive nature of the task, AI agents often host their models off-chain. Some agents rely on centralized cloud services such as AWS, while those that prefer decentralization use distributed computing networks such as Akash or ionet and Arweave for data storage.

由于任务的计算密集性，人工智能代理通常将其模型托管在链外。一些代理依赖 AWS 等集中式云服务，而那些喜欢去中心化的代理则使用 Akash 或 ionet 和 Arweave 等分布式计算网络进行数据存储。

Although the AI Agent model runs off-chain, the agent needs to interact with the on-chain protocol to execute smart contract functions and manage assets. This interaction requires a secure key management solution, such as an MPC wallet or a smart contract wallet, to process transactions securely. Agents can operate through

虽然AI Agent模型运行在链外，但Agent需要与链上协议交互来执行智能合约功能并管理资产。这种交互需要安全的密钥管理解决方案，例如 MPC 钱包或智能合约钱包，以安全地处理交易。代理商可以通过