Original title: A Deep Dive into Frameworks: A Sector we think Could Grow to $20b+ Original source: Deep Value Memetics
Original translation: Azuma, Odaily Planet Daily
Summary of key points
In this report, we discuss the development of several major frameworks in the field of Crypto & AI. We will examine the current four major frameworks - Eliza (AI16Z), G.A.M.E (GAME), Rig (ARC), ZerePy (ZEREBRO), and analyze their technical differences and development potential.
In the past week, we have analyzed and tested the above four frameworks, and the conclusions are summarized as follows.
· We believe Eliza (approximately 60% market share, with a market cap of about $900 million when the original author wrote, approximately $1.4 billion at the time of writing) will continue to dominate market share. Eliza's value lies in its first-mover advantage and accelerated adoption by developers, as evidenced by 193 contributors on GitHub, 1,800 forks, and over 6,000 stars, making it one of the most popular software libraries on GitHub.
· G.A.M.E (approximately 20% market share, with a market cap of about $300 million when the original author wrote, approximately $257 million at the time of writing) has developed very smoothly so far and is experiencing rapid adoption, as announced earlier by the Virtuals Protocol, with over 200 projects based on G.A.M.E built, daily request numbers exceeding 150,000, and a weekly growth rate exceeding 200%. G.A.M.E will continue to benefit from the explosion of VIRTUAL and is likely to become one of the biggest winners in this ecosystem.
· Rig (approximately 15% market share, with a market cap of about $160 million when the original author wrote, approximately $279 million at the time of writing) features a modular design that is very appealing and easy to operate, expected to dominate in the Solana ecosystem (RUST).
· Zerepy (approximately 5% market share, with a market cap of about $300 million when the original author wrote, approximately $424 million at the time of writing) is a more niche application specific to a fervent ZEREBRO community, and its recent collaboration with the ai16z community may produce some synergistic effects.
In the above statistics, 'market share' is calculated by considering market capitalization, development records, and the breadth of the underlying operating system terminal markets.
We believe that AI frameworks will become the fastest-growing sector in this cycle, with the current sector total market cap of about $1.7 billion likely to grow easily to $20 billion. Compared to the valuations of Layer 1 at its peak in 2021, this figure may still be conservative—at that time, many single projects were valued at over $20 billion. Although the frameworks mentioned above serve different terminal markets (chains/ecosystems), given our belief that this sector will grow overall, adopting a market cap-weighted approach may be the most prudent.
The four major frameworks
At the intersection of AI and Crypto, several frameworks have emerged aimed at accelerating AI development, including Eliza (AI16Z), G.A.M.E (GAME), Rig (ARC), and ZerePy (ZEREBRO). From open-source community projects to performance-focused enterprise solutions, each framework caters to different needs and philosophies in agent development.
In the table below, we list the key technologies, components, and advantages of each framework.
This report will first focus on what these frameworks are, the programming languages they use, their technical architectures, algorithms, and unique features with potential use cases. We will then compare each framework based on usability, scalability, adaptability, and performance, while discussing their strengths and limitations.
Eliza
Eliza is an open-source multi-agent simulation framework developed by ai16z, designed to create, deploy, and manage autonomous AI agents. Developed using TypeScript as the programming language, it provides a flexible and scalable platform for building intelligent agents that can interact with humans across multiple platforms while maintaining a consistent personality and knowledge.
The core features of this framework include: support for simultaneous deployment and management of multiple unique AI personalities via a multi-agent architecture; a role system for creating diverse agents using a role file framework; long-term memory and perceptible context memory management features through an advanced retrieval-augmented generation system (RAG). Furthermore, the Eliza framework provides smooth platform integration, allowing reliable connections with Discord, X, and other social media platforms.
In terms of communication and media capabilities for AI agents, Eliza is an excellent choice. In communication, the framework integrates with Discord's voice channel features, X functionality, Telegram, and direct API access for customized use cases. On the other hand, the framework's media processing capabilities have expanded to include reading and analyzing PDF documents, extracting and summarizing linked content, audio transcription, video content processing, image analysis, and dialogue summarization, effectively handling various media inputs and outputs.
Eliza provides flexible AI model support, allowing for local inference using open-source models, cloud-based inference with default configurations such as OpenAI and Nous Hermes Llama 3.1 B, and supports integration with Claude for handling complex queries. Eliza employs a modular architecture, featuring a wide-ranging action system, custom client support, and a comprehensive API, ensuring cross-application scalability and adaptability.
Eliza's use cases cover multiple domains, such as AI assistants related to customer support, community management, and personal tasks; social media roles like automated content creators and brand representatives; it can also serve as knowledge workers, acting as research assistants, content analysts, and document processors; and interactive roles like role-playing robots, educational mentors, and agent representatives.
Eliza's architecture is built around an agent runtime that seamlessly integrates with a role system (supported by model providers), a memory manager (connected to a database), and an action system (linked to platform clients). The framework's unique features include a plugin system that allows for modular functionality extensions, supporting multimodal interactions such as voice, text, and media, as well as compatibility with leading AI models like Llama, GPT-4, and Claude. With its versatility and robust design, Eliza is a powerful tool for developing AI applications across domains.
G.A.M.E
G.A.M.E is developed by the official Virtuals team, officially called 'Generative Autonomous Multimodal Entities Framework', designed to provide developers with APIs and SDKs to experiment with AI agents. The framework offers a structured way to manage AI agent behaviors, decision-making, and learning processes.
· The core components of G.A.M.E are as follows: first, the 'agent prompting interface' is the entry point for developers to integrate G.A.M.E into agents to obtain agent behaviors.
· The 'perceptual subsystem' initiates sessions by specifying parameters such as session ID, agent ID, user, and other relevant details. It synthesizes incoming messages into a format suitable for the 'strategic planning engine', serving as the sensory input mechanism for AI agents, whether in the form of dialogue or reactions. The core here is the 'dialogue processing module', responsible for handling messages and responses from agents and collaborating with the 'perceptual subsystem' to effectively interpret and respond to inputs.
· The 'strategic planning engine' collaborates with the 'dialogue processing module' and 'on-chain wallet operator' to generate responses and plans. The engine operates on two levels: as a high-level planner creating broad strategies based on context or goals; and as a low-level strategy, transforming those strategies into actionable policies, further subdivided into action planners (for specifying tasks) and plan executors (for executing tasks).
· A separate but critical component is the 'world context', which references the environment, world information, and game state, providing necessary context for agents' decisions. Additionally, the 'agent library' stores long-term attributes such as goals, reflections, experiences, and personalities, all of which collectively shape the agent's behavior and decision-making processes. The framework utilizes 'short-term working memory' and 'long-term memory processors'—short-term memory retains relevant information about previous actions, outcomes, and current plans; in contrast, long-term memory processors extract key information based on importance, recency, and relevance. This memory stores knowledge about the agent's experiences, reflections, dynamic personality, world context, and working memory, enhancing decision-making and providing a foundation for learning.
· To enhance layout, the 'learning module' retrieves data from the 'perceptual subsystem' to generate general knowledge, which is fed back into the system to optimize future interactions. Developers can input feedback on actions, game states, and sensory data through the interface to enhance AI agents' learning and improve their planning and decision-making capabilities.
The workflow begins with developers interacting through the agent prompting interface; the 'perceptual subsystem' processes input and forwards it to the 'dialogue processing module', which manages interaction logic; then, the 'strategic planning engine' formulates and executes plans based on this information, utilizing high-level strategies and detailed action planning.
Data from the 'world context' and 'agent library' informs these processes, while working memory tracks immediate tasks. Meanwhile, the 'long-term memory processor' stores and retrieves knowledge over time. The 'learning module' analyzes results and integrates new knowledge into the system, enabling the agent's behavior and interactions to continuously improve.
Rig
Rig is an open-source framework based on Rust, designed to simplify the development of large language model (LLM) applications. It provides a unified interface for interacting with multiple LLM providers (such as OpenAI and Anthropic) and supports various vector storage options, including MongoDB and Neo4j. The framework's modular architecture includes core components such as the 'provider abstraction layer', 'vector storage integration', and 'agent system', facilitating seamless interactions with LLMs.
Rig's primary audience includes developers building AI/ML applications using Rust, with a secondary audience being organizations seeking to integrate multiple LLM providers and vector storage into their Rust applications. The resource library is organized using a workspace-based structure and contains multiple crates, achieving scalability and efficient project management. Rig's main features include the 'provider abstraction layer', which standardizes the API for completing and embedding LLM providers through consistent error handling; the 'vector storage integration' component provides an abstract interface for multiple backends and supports vector similarity search; the 'agent system' simplifies LLM interaction, supporting retrieval-augmented generation (RAG) and tool integration. Additionally, the embedding framework provides batch processing capabilities and type-safe embedding operations.
Rig leverages multiple technical advantages to ensure reliability and performance. Asynchronous operations utilize Rust's asynchronous runtime to efficiently handle a large number of concurrent requests; the framework's inherent error handling mechanism improves recovery from failures in AI providers or database operations; type safety prevents compile-time errors, enhancing code maintainability; efficient serialization and deserialization processes facilitate handling data in formats like JSON, which are crucial for communication and storage in AI services; detailed logging and instrumentation further aid in debugging and monitoring applications.
The workflow in Rig begins with a client initiating a request, which flows through the 'provider abstraction layer' to interact with the corresponding LLM model; then, data is processed by the core layer, where agents can use tools or access vector storage for context. Responses are generated and refined through complex workflows like RAG, including document retrieval and context understanding, before being returned to the client. The system integrates multiple LLM providers and vector storage, adapting to model availability or performance changes.
Rig has a diverse range of use cases, including a question-answering system for retrieving relevant documents to provide accurate replies, document search and retrieval for efficient content discovery, and chatbots or virtual assistants providing context-aware interactions for customer service or education. It also supports content generation, capable of creating text and other materials based on learned patterns, making it a versatile tool for developers and organizations.
ZerePy
ZerePy is an open-source framework written in Python, designed to deploy agents on X using OpenAI or Anthropic LLMs. ZerePy is derived from a modular version of the Zerebro backend, allowing developers to launch agents with similar functionalities to the core features of Zerebro. While the framework provides a foundation for agent deployment, fine-tuning of models is necessary to produce creative outputs. ZerePy simplifies the development and deployment of personalized AI agents, particularly suited for content creation on social platforms, fostering an AI creative ecosystem targeted at art and decentralized applications.
The framework is built using Python, emphasizing the autonomy of agents, focusing on generating creative outputs, aligning with Eliza's architecture + partnerships. Its modular design supports memory system integration, facilitating the deployment of agents on social platforms. Its main features include a command-line interface for agent management, integration with X, support for OpenAI and Anthropic LLMs, and a modular connection system for enhanced functionality.
ZerePy's use cases encompass social media automation, where users can deploy AI agents for posting, replying, liking, and retweeting, thereby enhancing platform engagement. Additionally, it is suitable for content creation in areas like music, notes, and NFTs, making it an important tool for digital art and blockchain-based content platforms.
Horizontal Comparison
In our view, each of the frameworks above offers a unique approach to AI development, catering to specific needs and environments, which shifts the debate from whether these frameworks are competitors to whether each framework can provide unique utility and value.
· Eliza stands out with its user-friendly interface, especially suitable for developers familiar with JavaScript and Node.js environments. Its comprehensive documentation aids in setting up AI agents across various platforms, though its rich feature set may present a moderate learning curve. However, due to its use of TypeScript, Eliza is particularly well-suited for building agents embedded in web environments, as most front-end web infrastructure is built with TypeScript. The framework is known for its multi-agent architecture, capable of deploying diverse AI personality agents across platforms like Discord, X, and Telegram. Its advanced RAG system for memory management makes it especially suitable for building AI assistants for customer support or social media applications. While it offers flexibility, strong community support, and consistent cross-platform performance, it is still in its early stages, which may pose a learning curve for developers.
· G.A.M.E is designed for game developers, providing a low-code or no-code interface through APIs, making it accessible to those with lower technical proficiency in the gaming field. However, it focuses on game development and blockchain integration, which may present a steep learning curve for those without relevant experience. It excels in procedural content generation and NPC behavior, but is also limited by its niche and the additional complexity of blockchain integration.
· Rig may be less user-friendly due to the complexity of the Rust language, posing significant challenges to learning, but for those proficient in systems programming, it can offer intuitive interaction. Compared to TypeScript, Rust is known for its performance and memory safety. It features strict compile-time checks and zero-cost abstractions, essential for running complex AI algorithms. The efficiency and low-level control characteristics of the language make it ideal for resource-intensive AI applications. The framework adopts a modular and scalable design, providing high-performance solutions, making it very suitable for enterprise applications. However, for developers unfamiliar with Rust, using Rust can present a steep learning curve.
· ZerePy uses the Python language, providing higher usability for creative AI tasks. For Python developers, particularly those with an AI/ML background, the learning curve is lower, and due to ZEREBRO's popularity, strong community support is available. ZerePy excels in creative AI applications such as NFTs, positioning itself as a powerful tool in the digital media and art space. While it performs exceptionally in creativity, its scope of application is relatively narrow compared to other frameworks.
In terms of scalability, the four major frameworks compare as follows.
· Eliza has made significant progress since the update to version 2.0, introducing a unified message pipeline and an extensible core framework, achieving efficient cross-platform management. However, without optimization, managing this multi-platform interaction may present scalability challenges.
· G.A.M.E excels in real-time processing required for games, and its scalability can be managed through efficient algorithms and potential blockchain distributed systems, although it may be limited by specific game engines or blockchain networks.
· The Rig framework leverages the performance advantages of Rust for better scalability, inherently designed for high-throughput applications, which may be particularly effective for enterprise-level deployments, although achieving true scalability may require complex setups.
· ZerePy's scalability targets creative output and is supported by community contributions, but the framework's emphasis may limit its application in broader AI environments, with its scalability potentially tested by the diversity of creative tasks rather than user volume.
In terms of applicability, Eliza leads with its plugin system and cross-platform compatibility, followed by G.A.M.E in gaming environments and Rig for handling complex AI tasks. ZerePy shows high adaptability in creative domains but is less applicable in broader AI application areas.
In terms of performance, the test results for the four major frameworks are as follows.
· Eliza is optimized for quick interactions on social media, but its performance may vary when handling more complex computational tasks.
· G.A.M.E focuses on high-performance real-time interaction in gaming scenarios, utilizing efficient decision-making processes and potential blockchain for decentralized AI operations.
· Rig, based on Rust, can provide excellent performance for high-performance computing tasks, ideal for enterprise applications where computational efficiency is crucial.
· ZerePy's performance is targeted at the creation of creative content, with its metrics centered on the efficiency and quality of content generation, which may not be as universally applicable outside the creative domain.
Combining the aforementioned strengths and weaknesses, Eliza offers better flexibility and scalability, with its plugin system and role configuration providing strong adaptability, beneficial for cross-platform social AI interactions; G.A.M.E offers unique real-time interaction capabilities in gaming scenarios and introduces novel AI participation through blockchain integration; Rig's strengths lie in its performance and scalability, suitable for enterprise-level AI tasks, and emphasizes code simplicity and modularity to ensure the long-term health of projects; Zerepy excels in fostering creativity, leading in AI applications in digital art, and is supported by a vibrant community-driven development model.
In summary, each framework has its limitations. Eliza is still in its early stages, with potential stability issues and a long learning curve for new developers; G.A.M.E's niche focus may limit its broader application, and the introduction of blockchain adds complexity; Rig's learning curve is steeper due to the complexity of the Rust language, which may deter some developers; Zerepy's narrow focus on creative output may limit its application in other areas of AI.
Core comparison items sorting
Rig (ARC)
· Language: Rust, focusing on safety and performance.
· Use case: Focused on efficiency and scalability, making it an ideal choice for enterprise-level AI applications.
· Community: Less community-driven, more focused on technical developers.
Eliza (AI16Z)
· Language: TypeScript, emphasizing flexibility and community participation in Web3.
· Use case: Designed for social interaction, DAOs, and transactions, with a particular emphasis on multi-agent systems.
· Community: Highly community-driven, with extensive connections to GitHub.
ZerePy (ZEREBRO):
· Language: Python, more readily accepted by a broader group of AI developers.
· Use case: Suitable for social media automation and simpler AI agent tasks.
· Community: Relatively new, but expected to grow due to Python's popularity and support from ai16z contributors.
G.A.M.E (VIRTUAL, GMAE):
· Focus: Autonomous, adaptive AI agents that can evolve based on interactions in virtual environments.
· Use case: Most suitable for scenarios where agents need to learn and adapt, such as games or virtual worlds.
· Community: Innovative but still determining its positioning in the competition.
GitHub data growth trends
The above chart shows the changes in star data on GitHub since these frameworks were launched. Generally, GitHub stars serve as indicators of community interest, project popularity, and perceived value of the project.
· Eliza (Red Line): The chart shows a significant and stable increase in the number of stars for this framework, starting from a low base in July and surging in late November, now reaching 6,100 stars. This indicates a rapid increase in interest around this framework, attracting the attention of developers. The exponential growth suggests that Eliza has gained immense appeal due to its features, updates, and community participation, far exceeding that of other products, indicating strong community support and broader applicability or interest in the AI community.
· Rig (Blue Line): Rig is the most 'established' of the four major frameworks, with a modest but stable increase in stars, showing a noticeable rise in the past month. Its total stars have reached 1,700, but it remains on an upward trajectory. The steady accumulation of attention is due to ongoing development, updates, and a growing user base. This may reflect that Rig is a framework still building its reputation.
· ZerePy (Yellow Line): ZerePy just launched a few days ago, and the number of stars has grown to 181. It is important to emphasize that ZerePy needs more development to enhance its visibility and adoption, and its collaboration with ai16z may attract more contributors to its codebase.
· G.A.M.E (Green Line): The framework has a low number of stars, but it is noteworthy that it can be applied directly to agents in the Virtual ecosystem through the API, so there is no need to publish on GitHub. However, although the framework has only been publicly available for builders for just over a month, there are currently over 200 projects using G.A.M.E for building.
Expected upgrades of AI frameworks
Eliza's version 2.0 will include integration with the Coinbase agent toolkit. All projects using Eliza will gain support for future native TEE (Trusted Execution Environment), enabling agents to operate in a secure environment. The plugin registry is a feature that Eliza will soon launch, allowing developers to seamlessly register and integrate plugins.
Additionally, Eliza 2.0 will support automated anonymous cross-platform messaging. The Tokenomics white paper, expected to be released on January 1, 2025 (related proposals have already been announced), will positively impact the AI16Z token supporting the Eliza framework. ai16z plans to continue enhancing the framework's practicality and to leverage the efforts of its main contributors to attract high-quality talent.
The G.A.M.E framework provides no-code integration for agents, enabling the simultaneous use of G.A.M.E and Eliza within a single project, each serving specific use cases. This approach is expected to attract builders who focus on business logic rather than technical complexity. Although the framework has only been publicly available for over 30 days, it has made substantial progress with the team's efforts to attract more contributors. Every project launched on VirtuaI is expected to adopt G.A.M.E.
The Rig framework, powered by ARC tokens, has significant potential, although its growth is still in the early stages and the project contract plan to drive Rig adoption has only been live for a few days. However, high-quality projects are expected to emerge soon that pair with ARC, similar to the Virtual flywheel but focused on Solana. The Rig team is optimistic about collaborating with Solana, positioning ARC as Solana's Virtual. Notably, the team not only incentivizes new projects launched using Rig but also encourages developers to enhance the Rig framework itself.
ZerePy is a newly launched framework gaining significant attention due to its collaboration with ai16z (Eliza framework), attracting contributors from Eliza who are actively working to improve the framework. ZerePy enjoys fervent support driven by the ZEREBRO community and is opening new opportunities for Python developers who previously lacked space in the competitive AI infrastructure field. The framework is expected to play a significant role in AI creativity.
Original link