NVIDIA Clara™ Discovery is a collection of GPU-accelerated and optimized frameworks, tools, applications, and pre-trained models for computational drug discovery. Designed to support interdisciplinary workflows, Clara Discovery helps scientists and researchers bring drugs to market faster and create new opportunities for disease mechanism discovery.


MegaMoIBART, das Trainingsframework für große Sprachmodelle für den chemischen Bereich, erreicht die Molekülgenerierung im KI-Supercomputing-Maßstab mit hoher Validität und Einzigartigkeit.


Clara Discovery besteht aus einer Sammlung von Frameworks, Anwendungen und KI-Modellen, die eine GPU-beschleunigte Arzneimittelforschung ermöglichen und die Forschung in den Bereichen Genomik, Proteomik, Mikroskopie, virtuelles Screening, computergestützte Chemie, Visualisierung, klinische Bildgebung und natürliche Sprachverarbeitung (NLP) unterstützen. Sehen Sie sich diese Tools im NGC™-Katalog an, NVIDIAs Hub für GPU-optimierte Software.
CPU vs. GPU Full-Stack-Beschleunigung


Clara Discovery ist für den Einsatz auf dem NVIDIA DGX™ A100 konzipiert, dem weltweit fortschrittlichsten KI-System mit einer Leistung von fünf petaFLOPS. DGX A100 wurde speziell für alle großen beschleunigten Rechenlasten entwickelt und bietet Forschern die kürzeste Zeit bis zur Lösung und eine einheitliche, einfach zu implementierende Infrastruktur zur Unterstützung der nächsten Generation der Arzneimittelforschung.

Supported by deep learning and neural transformer networks.

GPU-enabled deep learning algorithms and transformer models will accelerate every phase of drug discovery. From training large language models (LLMs) with understanding of chemical space to molecular dynamics simulations, protein structure predictions, and generative drug discovery, scientists are re-exploring the ever-expanding chemical universe with novel deep learning technologies.
Credit: Mahendra awale, CC BY-SA 3.0 https://creativecommons.org/licenses/by-sa/3.0 , via Wikimedia Commons  

Transformer-based large language models operate at supercomputing scale with BioNeMo.

Announced at GTC, BioNeMo is an application framework and cloud service based on NVIDIA NeMo Megatron for training and deploying large-scale biomolecular transformer AI models at supercomputing scale. BioNeMo has pre-trained large language models and is tailored to the language of proteins, DNA, and the simplified molecular input lineage system (SMILES).

Accelerate key drug discovery applications.

Drug discovery requires many workflows - from exploring the chemical universe to predicting protein structures to analyzing drug candidates to simulating molecules. Drive breakthroughs with Clara Discovery's powerful tools available in the NVIDIA NGC Catalog™.

Chemoinformatics in detail

Transformer-based large language models create new opportunities for real-time exploration of the chemical universe. BioNeMo is a domain-specific framework based on NeMo Megatron for training and deploying biomolecular LLMs at supercomputing scale. It includes the MegaMolBART, ESM-1b, and ProtT5 transformer models.

MegaMolBART is a generative chemistry model trained with 1.4 billion molecules (SMILES strings) that can be used for a variety of chemoinformatics applications in drug discovery, such as reaction prediction, molecular optimization, and de novo small molecule generation.

For ProtT5 and ESM-1b, it has been shown that unsupervised pre-training can be used to generate learned embeddings that incorporate features for predicting protein structure, function, cellular location, water solubility, membrane binding, conserved and variable regions, and more.

Predict protein structures

Deep-learning-based approaches such as RELION enable high-throughput automation of cryoelectron microscopy (cryo-EM) to determine protein structures In RELION, an empirical Bayesian approach to cryo-EM analysis is implemented to make single or multiple 3D reconstructions as well as 2D class average more accurate.

To understand protein structures with atomistic detail, tools such as MELD can be used to infer structures from sparse, ambiguous, or imprecise data. MELD uses data in a physics-based Bayesian framework to improve protein structure determination.

Accelerate virtual screening

AI and accelerated computing can be used to search for a rigid target protein in millions of drug candidates. AutoDock is a growing collection of computational docking and virtual screening methods used in structure-based drug discovery and the study of fundamental mechanisms of biomolecular structure.

Support molecular dynamics simulations

GPU-based molecular dynamics frameworks can simulate the fundamental mechanisms of cells and calculate how strongly a particular drug binds to the desired target protein. Potentials acquired through machine learning that show promise for precision, energies, and forces at the quantum mechanical level fundamentally change molecular simulation.

Clara Discovery incorporates a variety of tools and frameworks for molecular simulation, including GROMACS, NAMD, Tinker-HP, VMD, TorchANI, and DeePMD-Kit.

Discover solutions for accelerated computing

Optimized for R&D in the pharmaceutical industry

Clara Discovery is designed to run on NVIDIA DGX™ A100, the world's most advanced AI system with five petaFLOPS of performance. DGX A100 is designed specifically for all large-scale accelerated computing workloads, providing researchers with the shortest time to solution and a unified, easy-to-deploy infrastructure to support the next generation of drug discovery.

view HPC Application Performance

view HPC Application Performance

view white paper on pharmaceutical industry PODs

learn more about Schrödinger's advanced computing platform

Learn more about NVIDIA DGX A100

Full stack acceleration

CPU vs. GPU: NVIDIA Clara Parabricks, Relion, Autodock-GPU, NVIDIA RAPIDS, Amber, NAMD, VMD, Gromacs, NVIDIA Clara Imaging, BERT Training



NVIDIA Clara AGX Developer Kit
The NVIDIA Clara AGX Developer Kit provides real-time AI and imaging for medical devices. Combining the efficient Jetson AGX Xavier Embedded ARM SoC, the powerful RTX 6000 GPU, and the 100 GbE connectivity of the ConnectX-6 SmartNIC with reference designs for AI ultrasound and endoscopy applications, Clara AGX provides an easy-to-use platform for developing a software-defined, AI-enabled medical device in real-time at the point of care.
NVIDIA Clara AGX Software
The NVIDIA Clara AGX SDK runs on the NVIDIA Clara AGX and Jetson platforms and provides developers with the ability to create end-to-end streaming workflows for medical imaging. It includes advanced examples for ultrasound video and endoscopy. An NVIDIA Developer Account is required to access the NVIDIA Clara AGX SDK.


Clara AGX Hardware
clara AGX Xavier Developer Kit (SKU #): Includes NVIDIA RTX graphics processor and NVIDIA ConnectX-6 network cards
Essential components
Discrete GPU
Xavier AGX 32 GB DRAM Module
CPU, GPU and I/O Processing
Mellanox ConnectX-6 Smart Interconnect
Xavier AGX 32 GB DRAM Module
Removable Storage
8-core Carmel ARM v8.2 64-bit CPU, 8MB L2 + 4MB L3
32GB 256-Bit LPDDR4x | 136.5GB/s
(RTX 6000)
4608-core Turing GPU with 576 Tensor Cores and 72 RT Cores
(AGX Xavier)
512-core Volta GPU with 64 Tensor Cores
GPU Memory
(RTX 6000)
GPU FP32 Performance
(RTX 6000)
(AGX Xavier)
32GB eMMC 5.1
(Removable SSD)
250 GB
350 W
(AGX Xavier)
Up to 2x 1000MP/sec
(RTX 6000)
Up to 1000MP/sec
(AGX Xavier)
Up to 2x 1500MP/sec
(RTX 6000)
Up to 1500MP/sec
1 x PCIe Gen4 x8 128 Gbps
1 x PCIe Gen4 x8 128 Gbps
(to AGX Xavier)
2 x USB 3.1 Gen 2
1 x USB 2.0
1 x 100 Gbps Ethernet QSFP
(To Mellanox)
1 x 10 Gbps Ethernet RJ45
(To Mellanox)
1 x 1 Gbps Ethernet RJ45
(to AGX Xavier)
1 x SD Card
1 x USB-C
Debug Interface
Video Input
Video Output
HDMI Out (from AGX Xavier)
DP Out (from RTX 6000)
How to get it

The Clara AGX Developer Kit is available exclusively to members of the NVIDIA Clara Developer Partner Program. Register for the Developer Partner Program through the application and an NVIDIA sales representative will contact you to determine next steps. Users can only sign up with a corporate or university email address.

Get the Clara AGX Developer Kit
How to install it
A host development PC

The development environment for Clara AGX requires a host development PC (not included with the Clara AGX Developer Kit).
How to develop them
The Clara AGX SDK runs on the Jetson platform and offers developers the ability to create end-to-end streaming workflows for medical imaging. It includes advanced examples for ultrasound video and endoscopy. An NVIDIA Developer Account is required to access the NVIDIA Clara AGX SDK.
SDK Features
  • NVIDIA RTX support - Docker and CUDA, TensorRT, dGPU activation.
  • NVIDIA Rivermax 100GbE Streaming - Rivermax transport protocol streams data over Ethernet directly into GPU GDDR DRAM with GPU Direct technology.
  • NVIDIA Clara Guardian Support - Develop AI applications to improve patient care and operational efficiency using everyday sensors such as cameras and microphones.
  • Windows device mode support - I/O driver enables Jetson AGX to operate in device mode on Windows 10 machines.
  • Endoscopy with NVIDIA DeepStream - Video in and inference for endoscopy and other video-based modalities
  • Sensor processing - support for serial camera interfaces
  • Reference application for AI endoscopy and ultrasound


Installing the Clara AGX software
with the SDK Manager
NGC Container Sammlung
Was ist NVIDIA Clara Discovery?

Clara Discovery is a collection of frameworks, applications, and AI models that enable GPU-accelerated computational drug discovery.

Drug development is an interdisciplinary endeavor. Clara Discovery can be used throughout the drug discovery process, combining accelerated computing, AI, and machine learning in genomics, proteomics, microscopy, virtual screening, computational chemistry, visualization, clinical imaging, and natural language processing.

What is included in this collection?

Clara Discovery is a growing collection of frameworks, applications and models that enable GPU-accelerated computational drug discovery. In particular, Clara Discovery supports genomics workflows with Clara Parabricks, cryoEM pipelines with Relion, virtual screening with Autodock, protein structure prediction with MELD, various third-party molecular simulation applications, pre-trained models and training frameworks from Clara Imaging, and Clara NLP with BioMegatron, pre-trained models from BioBert, and the NeMo training framework.


Clara Parabricks is a computational framework that supports genomic applications from DNA to RNA. It leverages NVIDIA's CUDA, HPC, AI, and data analytics stacks to build GPU-accelerated libraries, pipelines, and reference application workflows for primary, secondary, and tertiary analyses. Clara Parabricks is a complete portfolio of off-the-shelf solutions coupled with a toolkit to support the development of new applications that meet the needs of genomics labs.


Clara Imaging is a domain-optimized application framework for developers that includes a TensorFlow-based training framework with state-of-the-art pre-trained models to kick-start AI development using techniques such as Transfer Learning, Federated Learning, and AutoML. To enable faster creation of AI-enabled data, Clara Train includes APIs for AI-powered annotations that make any medical viewer AI-enabled.


Clara NLP is a collection of SOTA biomedical pre-trained language models and highly optimized pipelines for training NLP models on biomedical and clinical texts. With NeMo and BioMegatron, researchers and data scientists can build even more powerful NLP models on the large corpus of text data available to them.

Structure of the proteins

Relion RELION (REgularized LIkelihood OptimizatioN) implements an empirical Bayesian approach to electron cryo-microscopy (cryo-EM) analysis. In particular, it provides methods for refining single or multiple 3D reconstructions as well as 2D class averages. RELION is an important tool for the study of living cells.

MELD is a molecular simulation framework for determining protein structures by combining semi-reliable data with atomistic physical models through Bayesian inference.

Chemical Informatics

Cheminformatics is a demonstration of real-time exploration and analysis of a database of chemical compounds. Molecules are grouped based on chemical similarity and visualized in an interactive display. Users can explore regions of interest in chemical space in real time, generate molecules, and see the corresponding chemical structures and physical properties.

MegaMolBART is a seq2seq transformer model that understands chemistry and can be used for a variety of cheminformatics applications in drug discovery. The embeddings of its encoder can be used as features for predictive models. Alternatively, the encoder and decoder can be used together to generate new molecules by scanning the latent space of the model. MegaMolBART can be used in the real-time explorer or via the gRPC service.

Virtual Screening

Autodock is a growing collection of computational docking and virtual screening methods used in structure-based drug discovery and the study of fundamental mechanisms of biomolecular structure and function.

Molecular simulation

GROMACSis a molecular dynamics application for atomistic simulation. GROMACS is designed for the simulation of biomolecular systems such as proteins, lipids and nucleic acids. A variety of methods are supported.

NAMD is a parallel molecular dynamics code for high performance simulation of large biomolecular systems. NAMD uses the popular molecular graphics program VMD for simulation setup and trajectory analysis, but is also compatible with AMBER, CHARMM, and X-PLOR.

Tinker-HP is a CPU- and GPU-based massively parallel MPI package for long polarizable molecular dynamics simulations and polarizable QM/MM.

VMD is designed for modeling, visualization and analysis of biomolecular systems such as proteins, nucleic acids, lipid membranes, carbohydrate structures, etc. VMD provides a variety of graphical representations for visualization and coloring of molecular structures, as well as a powerful scripting language for post-processing and automation of visualization tasks.

Hardware requirements

Clara supports CUDA Compute Capability 6.0 and higher. This corresponds to GPUs of the Pascal, Volta, Turing and Ampere families.

Clara Train recommends NGC-compatible systems with NVIDIA Tesla V100 GPUs or NVIDIA Tesla T4 GPUs.

Driver requirements

Clara is based on NVIDIA CUDA 10.1.243, which requires NVIDIA driver version 418.xx. However, if you are working with Tesla (e.g. T4 or another Tesla card), you can use NVIDIA driver version 384.111+ or 410. You can also use driver version 396 for Tesla T4.


The end user license agreement is included with the product. The licenses are also included in the zip file of the model application. By dragging and using the Clara Train SDK container and downloading models, you accept the terms of these licenses.


Machen Sie mit Servervirtualisierung Ihren ersten Schritt auf dem Weg zur Cloud. Realisieren Sie dann in Ihrem eigenen Tempo Ihr vollständig virtualisiertes Software-Defined Datacenter, indem Sie Networking, Storage und Sicherheit virtualisieren, um virtuelle Rechenzentren aufzubauen. Vereinfachen Sie anschließend die Bereitstellung von IT-Ressourcen und Anwendungen, sodass Sie diese in wenigen Minuten zur Verfügung stellen können. Runden Sie die Entwicklung Ihrer Private Cloud schließlich mit einer Automatisierung des Managements ab, um optimale Performance, Kapazitätsnutzung und Compliance zu erzielen.
Dank der NVIDIA AI Enterprise-Suite und den fortschrittlichsten Grafikprozessoren und Datenverarbeitungseinheiten (DPUs) von NVIDIA können VMware-Kunden moderne, beschleunigte Workloads sicher zusammen mit bestehenden Unternehmensanwendungen auf NVIDIA-zertifizierten Systemen ausführen. Vom lokalen Rechenzentrum bis zur Hybrid-Cloud, von virtuellen Maschinen bis zu Containern.

WARUM sysgen?

As an NVIDIA Elite Partner with over 20 years of experience in High Performance Computing,
we are your partner for complex IT tasks in the software, hardware and medical solutions sector.

In addition to NVIDIA's flagship DGX H100, we naturally also offer other alternatives from the manufacturers. Whether NVIDIA DGX or Supermicro HGX systems in NVIDIA certified servers, single GPUs or industry specific server configurations-
we have the right solution for every budget.