M42-health

Models by this creator

🔮

med42-70b

m42-health

Total Score

142

med42-70b is an open-access clinical large language model (LLM) developed by M42 to expand access to medical knowledge. Built off LLaMA-2 and comprising 70 billion parameters, this generative AI system provides high-quality answers to medical questions. med42-70b was instruction-tuned on a dataset of ~250M tokens compiled from different open-access sources, including medical flashcards, exam questions, and open-domain dialogues. It is available for further testing and assessment as an AI assistant to enhance clinical decision-making and expand access to an LLM for healthcare use. Similar models include Meditron-70B from EPFL, a 70 billion parameter model adapted to the medical domain from Llama-2-70B, and BioMedLM from Stanford CRFM, a 2.7 billion parameter language model trained exclusively on biomedical abstracts and papers. Model inputs and outputs Inputs Text data**: med42-70b takes text-only data as input. Outputs Generative text**: The model generates text output in response to the input. Capabilities med42-70b can provide high-quality answers to medical questions, summarize patient records, aid in medical diagnosis, and handle general health Q&A. The model was trained to encode medical knowledge from reliable sources and can leverage this to assist healthcare professionals and expand access to medical information. What can I use it for? med42-70b is being made available for further testing and assessment as an AI assistant to enhance clinical decision-making and expand access to an LLM for healthcare use. Potential use cases include: Medical question answering Patient record summarization Aiding medical diagnosis General health Q&A Developers can use the model by following the specified formatting, including the `, and ` tags. Things to try With its medical domain knowledge and generation capabilities, med42-70b could be used to prototype a range of healthcare applications, such as virtual medical assistants, patient education tools, or clinical decision support systems. Developers should keep in mind the model's limitations and ensure thorough testing for safety and reliability before deploying in production environments.

Read more

Updated 4/28/2024