Jivi, an Indian healthcare AI startup co-founded by former BharatPe Chief Product Officer Ankur Jain and GV Sanjay Reddy, Chairman of Reddy Ventures, has achieved a remarkable feat.
The startup’s purpose-built medical Large Language Model (LLM), Jivi MedX, has secured the number one spot on the Open Medical LLM Leaderboard. This prestigious ranking places Jivi ahead of established giants like OpenAI’s GPT-4 and Google’s Med-PaLM 2, with an impressive average score of 91.65 across nine benchmark categories.
The Open Medical LLM Leaderboard, hosted by AI platform Hugging Face in collaboration with the University of Edinburgh and Open Life Science AI, evaluates medical-specific LLMs based on their performance in answering medical questions from various exams and research assessments. The evaluation spans medical entrance exams such as AIIMS and NEET in India, the US Medical License Exams (USMLE), and detailed assessments in clinical knowledge, medical genetics, and professional medicine.
“Jivi is revolutionizing primary healthcare through generative AI, making top-quality care accessible 24/7 at a fraction of the cost,” said Ankur Jain, Co-founder and CEO of Jivi. “Our mission at Jivi is to harness artificial intelligence to enhance patient care. Our platform accelerates diagnostics and ensures higher accuracy, enabling timely and precise treatment for all.”
Operating with a lean team of 20 members, including physicians, surgeons, AI engineers, and data scientists, Jivi is developing technology that aims to transform the accessibility, affordability, and quality of healthcare on a global scale.
“This is a monumental achievement for an Indian company. At Jivi, our mission is to make top-of-the-line healthcare available to everyone globally,” said GV Sanjay Reddy, Co-founder and Chairman of Jivi. “Our LLM being the best worldwide gives us immense pride and confidence as we prepare to bring Jivi to over a billion people.”
Jivi leverages its extensive proprietary medical dataset, comprising millions of medical research papers, journals, clinical notes, and other sources, to train its Jivi MedX LLM. This dataset is among the largest in the world. Jivi MedX was trained using an instruction fine-tuning algorithm called Odds Ratio Preference Optimization (ORPO), ensuring its superior performance.
As Jivi continues to push the boundaries of medical AI, its success on the global stage underscores the potential of Indian innovation in transforming healthcare worldwide.
