ai16z advisor Sam Gao has unveiled the technical whitepaper for Eliza, an open-source AI agent operating system.
Per a Jan. 14 X post from Gao, the white paper was written by Shaw, the founder of ElizaOS, alongside contributors from other related organizations.
According to the technical whitepaper, Eliza is described as a platform for deploying AI-powered agents that integrate directly with web3 technologies, including blockchain data interaction, smart contract operations, and decentralized applications. These agents are able to
“By doing so, it could pave the way for a more equitable world where the benefits of technological progress are more broadly and fairly distributed among humanity,” the paper adds.
To achieve this, Eliza has been developed using TypeScript, a widely used language in web3 and web development. As such, developers can integrate blockchain functionality into existing web applications and build dapps using familiar tools and frameworks.
Eliza’s modular architecture further enhances its adaptability by decoupling its structure into a core Runtime and four key components: Adapter (for data integration), Character (defining agent personalities), Client (managing message interactions), and Plugin (providing universal functionalities).
This design allows developers to freely add and customize AI agents without modifying the core Runtime.
Eliza is able to support a wide range of blockchain ecosystems, including Solana, Ethereum, and Ton, alongside AI model providers such as OpenAI, Llama, and Qwen. Its flexibility extends to platform integrations like Twitter, Discord, and Telegram, making it a powerful tool for deploying advanced web3 applications and AI functionalities.
The framework also supports plugins for image, video, and NFT generation, as well as blockchain analytics tools for portfolio management and decision-making.
However, the whitepaper notes that Eliza is still in its developmental phase and has identified workflow automation, runtime efficiency, and multi-language support as areas it still needs to improve upon. As its end goal, Eliza aims to evolve into a fully autonomous AI system capable of planning and executing tasks without human intervention.
As previously reported by crypto.news, Shaw revealed that Eliza v2 would integrate Hierarchical Task Networks (HTNs), a system designed to break down complex tasks into smaller, manageable components structured in a tree-like hierarchy. This will allow AI agents developed on Eliza to efficiently plan, adapt, and execute intricate, multi-step processes, even in uncertain or dynamic scenarios.
According to Elizas.world, a community-run dashboard, Eliza currently powers over 30 AI agents.
The launch of the technical whitepaper comes as AI agents continue to gain traction across various industries. Market research firm RootsAnalysis estimates the AI agent industry, valued at $5.29 billion in 2024, will grow at a projected compounded annual growth rate of 40% through 2035.
However, the crypto AI agent sector has faced some hiccups amid a broader market correction in the crypto space, with its value dropping 16% to $11.3 billion yesterday before recovering to $13.65 billion at press time.
