InternLM

Welcome to the InternLM organization. InternLM is mainly developed by Shanghai AI Laboratory. We keep open-sourcing high quality LLMs as well as a full-stack toolchain for development and application.

Models

• Intern-S1: a multimodal reasoning model with strong general capabilities and state-of-the-art performance on a wide range of scientific tasks.
• InternLM: a series of multi-lingual foundation models and chat models.
• InternLM-Math: state-of-the-art bilingual math reasoning LLMs.
• InternLM-XComposer: a vision-language large model (VLLM) based on InternLM for advanced text-image comprehension and composition.

Toolchain

• InternEvo: a lightweight framework for large-scale model pre-training and finetuning.
• XTuner: a toolkit for efficiently fine-tuning LLMs, supporting various models and fintuning algorithms.
• LMDeploy: a toolkit for compressing, deploying, and serving LLMs.
• Lagent: a lightweight framework that allows users to efficiently build LLM-based agents.
• AgentLego: a library of versatile tool APIs to extend and enhance LLM-based agents, compatible with Lagent, Langchain, etc.
• OpenCompass: a platform for large model evaluation, providing a fair, open, and reproducible benchmark.
• OpenAOE: an elegant and out-of-the-box chat UI for comparing mulitple models.

Applications

• HuixiangDou: a domain-specific assistant based on LLMs which can deal with complex techinical questions in group chats.
• MindSearch: an LLM-based multi-agent framework of web search engine.