CALMS : Context-Aware Language Model for Science

The software repository for the CALMS publication. CALMS is a retrieval and tool augmented large language model (LLM) to assist scientists, design experiments around and perform science using complex scientific instrumentation. When run with the open-source flag --hf, the models can be run on a local machine with sufficent ram or vram to contain the model. For running with OpenAI, the software is designed to interface with an internal Argonne API. This API call can be replaced with the standard OpenAI module in langchain (https://python.langchain.com/docs/integrations/llms/openai/) to replicate calls to GPT-3.5.