Bioptimus Launches M-Optimus, Revolutionizing Biological Modeling

Bioptimus has made a significant advancement in the field of biological modeling with the launch of its new platform, M-Optimus, on December 17, 2025, in Paris. This innovative model integrates multiple biological modalities, marking a transformational shift towards AI-powered virtual tissues and digital twins for patients. M-Optimus aims to provide a comprehensive framework for understanding biology in all its complexity, which is expected to enhance data-driven biomedical innovation.

Traditionally, biological research and AI models have operated in silos, focusing on isolated types of data such as genetic sequences or medical imaging. This segmented approach has limited the effectiveness of healthcare delivery. M-Optimus resolves these issues by offering a unified model that simulates various biological aspects, including cells, tissues, and patient data across diverse diseases and populations. The model combines multiple modalities, such as hematoxylin and eosin (H&E) stained histology images, bulk RNA sequencing, spatial transcriptomics, and clinical information, setting a new global standard in modeling biology.

The training of M-Optimus relied on one of the world’s largest proprietary datasets, encompassing millions of patients and over 50 organ types, sourced from hundreds of medical centers. This extensive training empowers the model to support a wide range of applications in the biomedical sector, from diagnostics to drug discovery and clinical practice.

Early Access Program for Pioneer Partners

Bioptimus is currently offering early access to M-Optimus for a select group of pioneering partners, including leading pharmaceutical companies who are integrating this model into their core research workflows. This program is designed to provide these organizations a competitive edge as they explore AI-driven biological discoveries.

Jean-Philippe Vert, CEO and co-founder of Bioptimus, emphasized the importance of this initiative: “With M-Optimus, we have successfully assembled the first critical components of our journey to crack the code of biology by combining multiple modalities at scale. This early access program is designed to partner with first-mover companies who share our vision: to translate this raw scientific power into tangible breakthroughs that ultimately improve patient outcomes and revolutionize the delivery of healthcare.”

Continuing Excellence with H-Optimus-1

While M-Optimus represents a new frontier, Bioptimus continues to lead in AI models for histopathology with its established model, H-Optimus-1. Recognized as a state-of-the-art solution, H-Optimus-1 has been trained on millions of whole-slide images, providing best-in-class performance in cancer grading and biomarker detection.

As the H-Optimus family of models approaches one million downloads, Bioptimus has partnered with Amazon SageMaker AI to enhance accessibility. This collaboration allows researchers, labs, and industry players to deploy the powerful pathology model directly into their secure cloud infrastructure, ensuring full data privacy and streamlined procurement processes.

Bioptimus’s advancements are supported by strategic partnerships with leading organizations such as AWS, HuggingFace, NVIDIA, Owkin, and Proscia. The company boasts a world-class team comprised of over 50% PhDs from top institutions like MIT, ETH, ENS, and TUM, alongside leadership from prominent organizations including Google Brain and DeepMind.

About Bioptimus: The company is recognized as a pioneer in AI-driven biotech, focusing on building a universal foundation model for biology. By merging advanced AI techniques with extensive, multimodal proprietary data, Bioptimus aims to create an integrative framework that connects all biological scales, from molecules to patients, delivering actionable insights that can significantly impact medical research and healthcare delivery.

For more information about Bioptimus and its innovations, visit: www.bioptimus.com.