ModelScope-Agent：使用开源大型语言模型构建可定制的代理系统

ModelScope-Agent: Building Your Customizable Agent System with Open-source Large Language Models

September 2, 2023

作者: Chenliang Li, Hehong Chen, Ming Yan, Weizhou Shen, Haiyang Xu, Zhikai Wu, Zhicheng Zhang, Wenmeng Zhou, Yingda Chen, Chen Cheng, Hongzhu Shi, Ji Zhang, Fei Huang, Jingren Zhou

cs.AI

摘要

大型语言模型（LLMs）最近展示了出色的能力，能够理解人类意图，进行推理，并设计类似规划的行为。为了进一步释放LLMs的力量以完成复杂任务，越来越多的趋势是构建代理框架，为LLMs（如ChatGPT）配备工具使用能力，以连接大量外部API。在这项工作中，我们介绍了ModelScope-Agent，这是一个通用且可定制的代理框架，用于实际应用，基于开源LLMs作为控制器。它提供了一个用户友好的系统库，具有可定制的引擎设计，支持在多个开源LLMs上进行模型训练，同时还能以统一的方式与模型API和常见API实现无缝集成。为了赋予LLMs工具使用能力，提出了一个全面的框架，涵盖了工具使用数据收集、工具检索、工具注册、内存控制、定制模型训练以及实际应用的评估。最后，我们展示了ModelScopeGPT，这是一个基于ModelScope-Agent框架的实际智能助手，能够连接开源LLMs与超过1000个公共AI模型以及ModelScope中的本地化社区知识。ModelScope-Agent库https://github.com/modelscope/modelscope-agent和在线演示https://modelscope.cn/studios/damo/ModelScopeGPT/summary 现已公开提供。

English

Large language models (LLMs) have recently demonstrated remarkable capabilities to comprehend human intentions, engage in reasoning, and design planning-like behavior. To further unleash the power of LLMs to accomplish complex tasks, there is a growing trend to build agent framework that equips LLMs, such as ChatGPT, with tool-use abilities to connect with massive external APIs. In this work, we introduce ModelScope-Agent, a general and customizable agent framework for real-world applications, based on open-source LLMs as controllers. It provides a user-friendly system library, with customizable engine design to support model training on multiple open-source LLMs, while also enabling seamless integration with both model APIs and common APIs in a unified way. To equip the LLMs with tool-use abilities, a comprehensive framework has been proposed spanning over tool-use data collection, tool retrieval, tool registration, memory control, customized model training, and evaluation for practical real-world applications. Finally, we showcase ModelScopeGPT, a real-world intelligent assistant of ModelScope Community based on the ModelScope-Agent framework, which is able to connect open-source LLMs with more than 1000 public AI models and localized community knowledge in ModelScope. The ModelScope-Agent libraryhttps://github.com/modelscope/modelscope-agent and online demohttps://modelscope.cn/studios/damo/ModelScopeGPT/summary are now publicly available.