Contribution to a conference proceedings

PUBDB-2025-05135

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png The Helmholtz Model Zoo: A Cloud-Based Platform for AI Model Sharing and Inference in the Helmholtz Association

Werners, H. W.DESY* ; Eren, E.DESY* ; Fuhrmann, P.DESY* ; Heuser, P.DESY*

2025
Zenodo

This record in other databases:  

Please use a persistent id in citations: doi:10.5281/ZENODO.17570188  doi:10.3204/PUBDB-2025-05135

Abstract: The Helmholtz Model Zoo (HMZ) is a cloud-based platform enabling seamless sharing and inference of deep learning models across the HelmholtzAssociation’s 18 research centers. By automating model deployment andproviding both web and programmatic interfaces, the HMZ lowers technicalbarriers to AI adoption in scientific research. Integrated with Helmholtzinfrastructure (Helmholtz ID authentication, dCache storage, DESY’s HPCcluster with NVIDIA L40S GPUs), the platform ensures secure, scalable inference while maintaining data sovereignty. NVIDIA Triton Inference Serverand Slurm manage GPU resources efficiently, supporting data-sets from gigabytes to terabytes. Virtual organizations enable fine-grained access controlfor specialized models. Launched in July 2025 in beta, the HMZ focuseson domain-specific applications, with future plans for model quality metricsand agentic capabilities.

Note: cc-by

---

Contributing Institute(s):

1. Informationstechnologie (IT)

Research Program(s):

1. 623 - Data Management and Analysis (POF4-623) (POF4-623)

Experiment(s):

1. No specific instrument

Appears in the scientific report 2025

---

Database coverage:
;

---