Major Changes in the Eliza Framework of AI16Z: From V1 to V2

Recently, the Eliza framework of AI16Z is about to undergo an important upgrade. By comparing the current version with the upcoming V2 version, we can clearly see the significant progress of the Eliza framework in architectural design and functional implementation.

Eliza Current Version Architecture

The current Eliza framework consists of three main levels:

1. Interface Layer: Responsible for interacting with external platforms such as Discord, Twitter, and Telegram, and providing API interfaces for other systems to call.

2. Core Functionality Layer: Includes the memory system ( that provides storage and document processing capabilities ), the behavior system ( that manages custom behaviors and action execution ), and the configuration system ( that manages various settings through configuration files ).

3. Model Layer: Supports various large language models such as OpenAI, Anthropic, Llama, and Grok, and provides a unified model interface abstraction.

This layered design gives Eliza a high degree of modularity and scalability.

Main Improvements of Eliza V2 Version

The alpha version of Eliza V2 is expected to be released in 2-3 weeks and will bring the following significant upgrades:

1. Interface Layer Improvement: Introduce a unified message bus to integrate the originally dispersed clients into a unified message processing system. Each platform will act as an adapter for the message bus, simplifying development and maintenance work, and providing a more consistent message processing experience.

2. Core Functionality Layer Improvements: Launch a unified proxy wallet that integrates the originally dispersed wallet configurations, providing unified multi-chain wallet management and simplifying cross-chain operations. Meanwhile, the core interface has been redesigned to enhance framework scalability and improve plugin integration mechanisms.

3. Model Layer Improvement: Introduce a model registry mechanism to support dynamic registration and overriding of models. Divide models into two main categories: LLMs and Embeddings, to achieve a more flexible model invocation method.

4. Other improvements: update the community plugin system, improve plugin management; achieve 100% test coverage to ensure framework quality.

Advantages of V2 Architecture

The new architecture of Eliza V2 is more modular and unified, with clearer interactions between components, laying a better foundation for future expansion. The unified message bus and proxy wallet will greatly simplify the development process and improve overall system efficiency. The improved plugin system and core framework design will provide developers with greater flexibility.

This upgrade not only enhances the technical strength of the Eliza framework but also creates new development opportunities for function expansion plugins such as AgentTank. With the release of version V2, we can expect to see more innovative applications emerging on the Eliza platform.