From federated models to drug discovery decisions
We deliver drug discovery AI models through secure, local applications.
Secure, local inference: Run models in your environment so data, queries, and outputs stay in-house.
In-house benchmarking: Compare model versions on curated internal datasets to understand reliability.
Customization: Fine-tune models for your targets and chemotypes, then integrate results via GUI or API.
Federated data networks that create superior models
We power the industry’s largest federated data networks spanning multiple pharmaceutical organizations.
Join the AI Structural Biology (AISB) Network
Build and use better protein-complex prediction models
Join the ADMET Network
Design more informative batches and use experimental capacity better in DMTA cycles
Join the Antibody Developability Network
Advance antibody R&D with purpose-built datasets and federated AI training
Designed for drug discovery teams
Drug discovery teams need to understand when models are reliable, how they behave on their own data, and how to adapt them as programs evolve. Apheris is designed for this reality: secure local use, benchmarking on in-house datasets, controlled customization, and access to models trained across industry data — all integrated into existing discovery workflows.
Powering the AISB Network to Revolutionize AI Drug Discovery
Apheris provides the tech layer for the Artificial Intelligence Structural Biology (AISB) Network, an unprecedented collaboration among AbbVie, Astex, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, Johnson & Johnson, Sanofi and Takeda aimed at transforming AI drug discovery. State-of-the-art AI models are trained and evaluated on unique data from multiple biopharma companies without exposing proprietary information.
Trusted by our customers and partners
What our customers and partners say about us
Through the AISB Network we can collaborate on data with other pharma partners, exploring the hypothesis that each of our internal data sets will be highly complementary when training AI models. The result could be transformative in how we advance AI-driven drug discovery to develop better medicines faster.
Fine-Tune Co-Folding Models on Your Own Data
ApherisFold can be used independently of any federated data network. Teams run co-folding models locally, benchmark models on in-house data, and fine-tune models on own, proprietary data within their own environment. Federated networks extend fine-tuning of models across organizations.
How companies engage with Apheris
Use applications locally
Deploy and run models in your environment for secure inference, benchmarking, and customization.
Join a federated data network
Participate in industry-wide data networks to improve model performance using cross-company data — without sharing.
Build your own network
Create a federated collaboration across partners, sites, or data sources using Apheris’ federated infrastructure.
