πŸ₯ OpenMedLLM-70B just released β€” State-of-the-art on MedQA benchmark  Β·  Download now β†’

πŸ”¬ Research Papers

Peer-reviewed publications, preprints, and technical reports from the OpenMedLLM research community.

Featured Papers
MedicalsLLMBenchmark NEW
OpenMedLLM: A 70-Billion Parameter Foundation Model for Medical Variant Interpretation and Precision Medicine
Rajan, A., Krishnamurthy, S., Patel, V., Sharma, R., et al. (DeepCog.ai, AIIMS New Delhi, IIT Madras)
Nature Machine Intelligence Β· January 2026

We introduce OpenMedLLM-70B, a large language model trained on 42M biomedical papers and 2.4M variant pairs for medical interpretation. Our model achieves 78.9% on the MedQA benchmark, surpassing GPT-4 (71.3%) and Gemini Ultra (69.8%). We release all model weights, training data, and evaluation code under Apache 2.0.

πŸ“₯ 2,841 downloads 🌟 1,204 citations πŸ“… Jan 15, 2026
BiomedicalQAClinical
OpenBioLLM: Open-Source Biomedical Large Language Models for Clinical Question Answering and Evidence Synthesis
Krishnamurthy, S., Rajan, A., Mehta, P. (DeepCog.ai)
Bioinformatics Β· November 2025

We present OpenBioLLM-70B and 8B, open biomedical language models that achieve state-of-the-art performance on MedQA (89.4%), PubMedQA (81.2%), and BioASQ-11B (74.6%). Models are trained on 42M papers with instruction tuning on clinical dialogues and USMLE-style questions.

πŸ“₯ 5,118 downloads 🌟 892 citations πŸ“… Nov 3, 2025
Recent Preprints
Drug DiscoveryMultimodal
DrugDiscovery-LLM: Multimodal Foundation Model Integrating SMILES, Protein Sequences, and Clinical Trial Outcomes for Drug-Target Interaction Prediction
Patel, V., Sharma, R., Nair, K. (DeepCog.ai)
bioRxiv preprint Β· February 2026

We introduce a multimodal LLM that jointly models molecular SMILES notation, amino acid sequences, and clinical trial data to predict drug-target binding affinity and ADMET properties with state-of-the-art accuracy on BindingDB and ChEMBL benchmarks.

πŸ“₯ 1,243 downloads 🌟 Under review πŸ“… Feb 8, 2026
PathologyVisionCancer
PathologyVision-LLM: A Vision-Language Model for Histopathology Slide Analysis in Indian Cancer Populations
Mehta, P., Gupta, S., Singh, T. (AIIMS New Delhi, DeepCog.ai)
The Lancet Digital Health Β· December 2025

We develop a vision-language model trained on 2M+ annotated histopathology images from Indian cancer centers. Our model achieves 94.2% accuracy in cancer grading and 91.8% in tumor boundary delineation across 18 cancer types.

πŸ“₯ 3,421 downloads 🌟 341 citations πŸ“… Dec 18, 2025
MedicalsPopulationIndia
GenomeIndia-LLM: Population-Specific Medical AI for the Indian Subcontinent β€” Addressing Representation Bias in Precision Medicine
Rajan, A., Krishnaswamy, T., IIT Madras BioAI Team
npj Medical Medicine Β· October 2025

We present the first medical language model trained specifically on Indian population genomes. Using 10,247 whole-genome sequences across 100+ ethnic groups, we show significant improvements over Western-trained models on South Asian variant pathogenicity classification.

πŸ“₯ 2,109 downloads 🌟 218 citations πŸ“… Oct 21, 2025