Analyzing AI Frameworks: From Intelligent Agents to Decentralization Exploration

Introduction

Recently, the narrative of the combination of AI and cryptocurrency has developed rapidly. Market attention has shifted to technology-driven "framework-type" projects, which have generated multiple projects with market capitalizations exceeding one hundred million and even one billion in just a few weeks. These projects have also given rise to a new asset issuance model: issuing tokens based on GitHub repositories, where Agents developed from the framework can also issue tokens. Based on the framework, with Agents as upper-layer applications, a model similar to an asset issuance platform is forming, which is actually a unique infrastructure model of the AI era. This article will start from the concept of frameworks and interpret the significance of AI frameworks in the cryptocurrency field in conjunction with personal reflections.

1. Framework Concept

AI frameworks are foundational development tools or platforms that integrate pre-built modules, libraries, and tools to simplify the process of building complex AI models. These frameworks typically include functionalities for data processing, model training, and prediction. In simple terms, frameworks can be understood as the operating systems of the AI era, similar to Windows and Linux in desktop systems, or iOS and Android in mobile devices. Each framework has its own strengths and weaknesses, allowing developers to choose based on their needs.

Although the "AI framework" is a new concept in the cryptocurrency field, its development has a history of nearly 14 years. There are mature frameworks available in the traditional AI field, such as Google's TensorFlow and Meta's PyTorch. The framework projects emerging in cryptocurrency are designed to meet the large demand for Agents under the current AI boom and extend to other fields, forming AI frameworks in different subfields.

1.1 Eliza

Eliza is a multi-Agent simulation framework developed by a16z, used for creating, deploying, and managing autonomous AI Agents. Developed based on TypeScript, it has good compatibility and API integration capabilities.

Eliza mainly targets social media scenarios, supporting multi-platform integration, including Discord, Twitter/X, Telegram, etc. In terms of media content processing, it supports functions such as PDF analysis, link content extraction, audio transcription, video processing, image analysis, and more.

The current use cases supported by Eliza mainly include:

1. AI Assistant Applications
2. Role of Social Media
3. Knowledge Workers
4. Interactive Roles

Eliza supports multiple models, including open-source model local inference and cloud-based inference.

1.2 G.A.M.E

G.A.M.E is an auto-generating and managing multimodal AI framework launched by Virtual, primarily aimed at designing intelligent NPCs in games. The framework is characterized by its usability for low-code or even no-code users, who can participate in Agent design simply by modifying parameters.

The core design of G.A.M.E adopts a modular architecture where multiple subsystems work collaboratively, including the Agent prompt interface, perception subsystem, strategic planning engine, world context, dialogue processing module, and various other components.

The framework mainly focuses on the decision-making, feedback, perception, and personality of agents in virtual environments, suitable for gaming and metaverse scenarios.

1.3 Rig

Rig is an open-source tool written in Rust, designed to simplify the development of large language model applications (LLM). It provides a unified operating interface, making it convenient for developers to interact with multiple LLM service providers and vector databases.

The core features of Rig include:

1. Unified Interface
2. Modular Architecture
3. Type Safety
4. High Performance

Rig is suitable for building question-and-answer systems, document search tools, chatbots, and scenarios that support content creation.

1.4 ZerePy

ZerePy is an open-source framework based on Python, designed to simplify the process of deploying and managing AI Agents on the X( platform before Twitter). It inherits the core functionalities of the Zerebro project but adopts a more modular and extensible design.

ZerePy provides a command line interface, supports large language models from OpenAI and Anthropic, directly integrates with X platform API, and plans to add a memory system in the future.

2. Similarities with the Development Path of the BTC Ecosystem

The development path of AI Agents has many similarities with the recent BTC ecosystem. The development of the BTC ecosystem can be summarized as: BRC20 - multi-protocol competition - BTC L2 - BTCFi. For AI Agents, it is: GOAT/ACT - social Agents/analytical AI Agents - framework competition. In the future, infrastructure projects focusing on the Decentralization and security of Agents may become the main theme of the next stage.

The AI framework project provides a new infrastructure development approach. Compared to Memecoin issuance platforms and inscription protocols, the AI framework resembles a future public chain, while Agent resembles a future Dapp. Future debates may shift from the EVM versus heterogeneous chain dispute to framework disputes, with the key issues being how to achieve Decentralization or chaining, and the significance of developing AI frameworks on the blockchain.

3. The Significance of On-Chain

The combination of blockchain and AI needs to address the issues of its meaning and value. Drawing on the successful experiences of DeFi, the reasons supporting the agent chainization may include:

1. Reduce usage costs, improve accessibility and choice
2. Provide blockchain-based security solutions
3. Achieve a unique blockchain financial model
4. Achieve transparent and traceable reasoning, enhancing interoperability.

4. Potential of the Creative Economy

AI framework projects may offer entrepreneurial opportunities similar to the GPT Store in the future. A framework that simplifies the Agent construction process and provides a combination of complex functionalities could gain an advantage, creating a more interesting Web3 creative economy than the GPT Store.

Web3 can address the shortcomings of the unfair policies of Web2 giants in terms of demand and economic systems, introducing a community economy that makes Agents more complete. The creative economy of Agents will provide opportunities for ordinary people to participate, and future AI Memes may be smarter and more interesting than the Agents on existing platforms.