
A lightweight open-source framework for building large language model (LLM)-based agents.
Lagent is an open-source framework developed by InternLM that enables users to efficiently create agents powered by large language models (LLMs). Inspired by PyTorch's design philosophy, Lagent offers a clear and intuitive workflow, allowing users to focus on creating layers and defining message passing between them in a Pythonic manner. The framework supports various LLMs, including API-based models like GPT-3.5/4 and open-source models such as LLaMA 2 and InternLM. Lagent also provides typical tools to augment LLMs, facilitating the development of complex multi-agent applications.
50%
Loading Community Opinions...