> For the complete documentation index, see [llms.txt](https://brindha.gitbook.io/mylearning/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://brindha.gitbook.io/mylearning/tools/internlm.md).

# InternLM

**What it is**

The InternLM family is a collection of open-source large language models developed by the Shanghai AI Laboratory, available in parameter sizes of 1.8B, 7B, and 20B. These models are trained on trillions of high-quality tokens and feature support for long context windows up to 1 million tokens in some versions, with enhanced reasoning capabilities particularly in mathematics and improved tool usage. 

**Collaborators**

InternLM is a multilingual large language model jointly developed by Shanghai AI Lab and SenseTime, in collaboration with the Chinese University of Hong Kong, Fudan University, and Shanghai Jiaotong University. 

**InternLM2.5 improvements**

InternLM2.5's reasoning capability has been significantly enhanced, surpassing the Llama3-70B model in some dimensions. On the MATH evaluation set, it achieved a 100% performance improvement over its predecessor. 

**Specialized variants**

The range includes specialized models like InternLM-XComposer, which excels in tasks involving text-image comprehension and composition. 

**Full open-source ecosystem**

Shanghai AI Laboratory has pioneered a full-chain open-source tool system covering data processing, pre-training, fine-tuning, deployment, evaluation, and application. Core tools including training framework XTuner, deployment framework LMDeploy, evaluation framework OpenCompass, and document parsing tool MinerU have all been fully open-sourced, forming a community with hundreds of thousands of developers. 

**License:** Apache 2.0.