AI company Tevogen.AI broadens partnerships with Databricks and Microsoft to construct the beta variant of PredicTcell Model, centering on oncology research.
Tevogen Bio Holdings Inc. (TVGN) has announced an expansion of its collaboration with Microsoft (MSFT) and Databricks to build the beta version of its PredicTcell model. The PredicTcell model, a foundational AI model developed by Tevogen.AI, is intended to accelerate cancer immunotherapy development.
The alpha PredicTcell model, developed in collaboration with Microsoft and Databricks, has already been completed, according to Tevogen's press materials. The company is now focusing on building a beta version of the model, with a specific focus on oncology.
The beta development is being powered on the Databricks Data Intelligence Platform and is backed by the engineering teams from Databricks and Microsoft. The aim is to extend the model from infectious-disease targets to cancer targets, with the explicit goal of accelerating cancer immunotherapy discovery and improving prediction of immunologically active peptides for cancer applications.
Tevogen.AI's recent international patent application (WO 2025/129197) outlines novel machine learning systems for predicting immunologically active peptides. This patent application is related IP underpinning PredicTcell’s development, according to Tevogen.
The PredicTcell model's development builds on Tevogen.AI's published international patent application. The model is designed to shorten target analysis timelines in immunotherapy research and improve candidate selection for T-cell–directed therapies, potentially reducing time and cost in drug development.
The oncology beta aims to increase model applicability to cancer neoantigens and other tumor immunogenic peptides. If validated, this could accelerate antigen selection and preclinical design for cancer vaccines or T-cell therapies, which is Tevogen's stated objective.
Tevogen is also planning to build enhanced analytics and visualization tools around PredicTcell to support its internal R&D teams and to enable broader platform use and interpretation. The company is also exploring external market and commercialization opportunities for PredicTcell, such as licensing or AI-as-a-service, so it could become a revenue stream beyond Tevogen's internal drug programs.
However, it's important to note that the announcements do not publish independent validation data, performance metrics of the alpha vs. beta, regulatory strategy details, nor formal third-party benchmarks of PredicTcell performance as of the August 2025 releases. Timelines for beta completion, prospective validation studies, integration into specific clinical programs, or commercial launch plans were not specified in the published release summaries.
The success of Tevogen may depend on its ability to manage its growth and expand operations, as well as its ability to develop and maintain effective internal controls. Economic conditions, competition, regulatory lawsuits or proceedings, and risks related to the ability to develop, license, or acquire new therapeutics could all impact Tevogen's growth and success.
Nevertheless, the development of the PredicTcell model is a significant step forward in the field of oncology, which is considered one of the most impactful areas for AI in drug discovery. The model's potential to accelerate cancer immunotherapy development could have far-reaching implications for the treatment of cancer.
References
- Tevogen Bio Holdings Inc. (2025, August). Tevogen.AI Announces Expanded Collaboration with Microsoft and Databricks to Build the Beta Version of PredicTcell. [Press release]
- Tevogen Bio Holdings Inc. (2025, August). Tevogen.AI Reports Completion of Alpha PredicTcell Model. [Press release]
- Tevogen Bio Holdings Inc. (2025, August). Tevogen.AI Expands Collaboration with Microsoft and Databricks to Build the Beta Version of PredicTcell. [News summary]
- World Intellectual Property Organization (2025). International patent application WO 2025/129197: Tevogen Bio Holdings Inc. Title: Novel machine learning systems for predicting immunologically active peptides. Abstract.
- Microsoft Corporation (2025). Databricks and Tevogen.AI Collaborate to Advance PredicTcell Program. [Press release]
- The PredicTcell model, an AI model developed by Tevogen.AI, focuses on extending its scope from infectious diseases to cancer, with the intention of improving health-and-wellness outcomes for medical-conditions such as cancer by accelerating immunotherapy discovery and predicting immunologically active peptides.
- The development of PredicTcell also employs technology such as artificial-intelligence and relies on collaboration with Microsoft and Databricks, with the ultimate goal of potentially reducing time and cost in drug development for various health-and-wellness concerns, including cancer.
- As part of its growth strategy, Tevogen is considering external market opportunities for PredicTcell, such as licensing or AI-as-a-service, which could contribute to science and technology advancements and generate revenue beyond Tevogen's internal drug programs.
