RestGPT

LLM-based Autonomous Agents Controlling Real-World Applications via RESTful APIs. This work aims to construct large-scale autonomous agents, RestGPT, based on language models to control real-world applications such as movie databases and music players. To achieve this goal, we integrate LLMs with RESTful APIs and address practical challenges in planning, API invocation, and response parsing. To thoroughly evaluate RestGPT's performance, we introduce RestBench—a high-quality benchmark comprising two real-world scenarios with human-annotated instructions and golden solution paths.

RestGPT is an autonomous agent based on large language models, designed to control real-world applications via RESTful APIs, addressing practical challenges such as planning, API calls, and response parsing.