[{"title":"MONAI/Llama3-VILA-M3-13B","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/MONAI/Llama3-VILA-M3-13B","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"An advanced multimodal model that analyzes MRI scans to provide structured clinical reasoning and detailed explanations for complex neurological or musculoskeletal findings.","document_title":"MONAI/Llama3-VILA-M3-13B","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"nvidia/NV-Segment-CT","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Segment-CT","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"This 3D foundation model provides out-of-the-box automatic and interactive point-click segmentation for 132 anatomical classes in computed tomography (CT) scans.","document_title":"nvidia/NV-Segment-CT","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"nvidia/NV-Segment-CTMR","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Segment-CTMR","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Expanding on its predecessor, this model automates the detailed segmentation of over 345 anatomical and pathological classes across both CT and magnetic resonance imaging (MRI) modalities.","document_title":"nvidia/NV-Segment-CTMR","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"nvidia/NV-Generate-CT","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Generate-CT","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"This 3D latent diffusion model synthesizes high-resolution, realistic 3D CT volumes alongside paired segmentation masks to help augment limited medical training datasets.","document_title":"nvidia/NV-Generate-CT","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"nvidia/NV-Reason-CXR-3B","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Reason-CXR-3B","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"A specialized vision-language model designed for medical reasoning and interpretation of chest X-ray images, with detailed explanations. The model combines visual understanding with medical reasoning capabilities, enabling healthcare professionals to access comprehensive analyses and engage in follow-up discussions about radiological findings.","document_title":"nvidia/NV-Reason-CXR-3B","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Riva | Speech and Translation AI","featured":false,"x_formats":["webpage"],"document_url":"https://www.nvidia.com/en-us/ai-data-science/products/riva/","technologies":["Riva"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"NVIDIA® Riva is a collection of GPU-accelerated microservices for building real-time, customizable speech AI functionality into your digital health applications.","document_title":"Riva | Speech and Translation AI","learning_level":"Technical - Beginner","x_content_types":["Overview"]},{"title":"NV-Generate-MR-Brain","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Generate-MR-Brain","technologies":["MONAI"],"document_date":"2026-03-12T04:00:00.000Z","short_summary":"This generative model synthesizes high-fidelity, 3D T1-weighted brain MRI volumes from noise, providing a powerful tool for creating diverse synthetic datasets to train other medical imaging algorithms.","document_title":"NV-Generate-MR-Brain","learning_level":"Technical - Advanced","x_content_types":["Samples"]},{"title":"nvidia/NV-Generate-MR","featured":false,"x_formats":["code"],"document_url":"https://huggingface.co/nvidia/NV-Generate-MR","technologies":["MONAI"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Similar to the CT variant, this generative AI model creates diverse, high-quality synthetic 3D MRI scans across various contrast modalities to supplement research data.","document_title":"nvidia/NV-Generate-MR","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"pangenome_aware_deepvariant","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_pangenome_aware_deepvariant.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Pangenome-aware DeepVariant is an enhanced version of Google's DeepVariant that leverages pangenome reference graphs (GBZ files) to improve variant calling accuracy, particularly in complex and highly variable genomic regions.","document_title":"pangenome_aware_deepvariant","learning_level":"Technical - Intermediate","x_content_types":["Documentation"]},{"title":"minimap2","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_minimap2.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Sequence alignment tool to align DNA or RNA sequencing reads to a reference genome. Often used for long-read technology from ONT or PacBio.","document_title":"minimap2","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"haplotypecaller","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_haplotypecaller.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Tool from Genome Analysis Toolkit (GATK) to detect germline/genetic variants.","document_title":"haplotypecaller","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"starfusion","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_starfusion.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Tool designed to detect gene fusion transcripts using RNA-Seq data–particularly from Illumina sequencing platforms.","document_title":"starfusion","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"mutectcaller","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_mutectcaller.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Tool from Genome Analysis Toolkit (GATK) for somatic variants.","document_title":"mutectcaller","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"NVFlare: NVIDIA Federated Learning Application Runtime Environment","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/NVFlare","technologies":["NVFLare"],"document_date":"2026-03-09T04:00:00.000Z","short_summary":"NVFlare (NVIDIA Federated Learning Application Runtime Environment) is an open-source, domain-agnostic Python SDK that provides a secure, extensible framework for collaborative computing, enabling researchers to adapt existing machine learning and data science workflows to a federated paradigm without centralizing sensitive data.","document_title":"NVFlare: NVIDIA Federated Learning Application Runtime Environment","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Clara Parabricks Workflows: GPU-Accelerated Genomic Analysis Pipelines","featured":false,"x_formats":["code"],"document_url":"https://github.com/clara-parabricks-workflows","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"clara-parabricks-workflows is a collection of GPU-accelerated genomic analysis pipelines, written in languages like WDL and Nextflow, designed to run NVIDIA Parabricks secondary analysis tools at high speed across various on-premises and cloud environments.","document_title":"Clara Parabricks Workflows: GPU-Accelerated Genomic Analysis Pipelines","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Single Cell Analysis Blueprint by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/single-cell-analysis","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Investigate, understand, and interpret single-cell data in minutes, not days, by leveraging RAPIDS-singlecell, powered by NVIDIA CUDA-X Data Science (RAPIDS).","document_title":"Single Cell Analysis Blueprint by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"alphafold2-multimer Model by DeepMind","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/deepmind/alphafold2-multimer","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Predicts the 3D structure of a protein from its amino acid sequence, specifically optimized for predicting the structures of multimeric protein complexes.","document_title":"alphafold2-multimer Model by DeepMind","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"deepsomatic","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_deepsomatic.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Google's deep-learning based variant caller for somatic variants (i.e., non-inherited diseases like cancer).","document_title":"deepsomatic","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"giraffe (vg giraffe + GATK)","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_giraffe.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Built by UCSC, a short-read mapping tool for pangenome graph alignment.","document_title":"giraffe (vg giraffe + GATK)","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"fq2bam_meth","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_fq2bam_meth.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Accurate alignment of bisulfite-treated DNA sequencing reads (BS-Seq) often used in methylation.","document_title":"fq2bam_meth","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"msa-search Model by Colabfold","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/colabfold/msa-search","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Generates a multiple sequence alignment from a query sequence by performing a high-speed protein sequence database search.","document_title":"msa-search Model by Colabfold","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"genmol Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/genmol-generate","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Executes fragment-based molecular generation via discrete diffusion to optimize molecules for user-defined objectives.","document_title":"genmol Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"openfold2 Model by Openfold","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/openfold/openfold2","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Predicts the 3D structure of a protein from its amino acid sequence, multiple sequence alignments, and templates.","document_title":"openfold2 Model by Openfold","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"llama-3.3-nemotron-super-49b-v1.5 Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/llama-3_3-nemotron-super-49b-v1_5","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Llama-3.3-Nemotron-Super-49B-v1.5 is a reasoning model that is post trained for reasoning, human chat preferences, and agentic tasks, such as RAG and tool calling.","document_title":"llama-3.3-nemotron-super-49b-v1.5 Model by NVIDIA","learning_level":"Technical - Intermediate","x_content_types":["Demo"]},{"title":"llama-3.1-nemotron-ultra-253b-v1 Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/llama-3_1-nemotron-ultra-253b-v1","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Llama-3.1-Nemotron-Ultra-253B-v1 is a large language model (LLM) which is a derivative of Meta Llama-3.1-405B-Instruct.","document_title":"llama-3.1-nemotron-ultra-253b-v1 Model by NVIDIA","learning_level":"Technical - Intermediate","x_content_types":["Demo"]},{"title":"parakeet-1.1b-rnnt-multilingual-asr Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/parakeet-1_1b-rnnt-multilingual-asr","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"The model transcribes 25 languages in upper case and lower case alphabets along with punctuations, spaces, and apostrophes.","document_title":"parakeet-1.1b-rnnt-multilingual-asr Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"nemotron-3-nano-30b-a3b Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/nemotron-3-nano-30b-a3b","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Nemotron-3-Nano-30B-A3B is a large language model (LLM) trained from scratch and designed as a unified model for both reasoning and non-reasoning tasks.","document_title":"nemotron-3-nano-30b-a3b Model by NVIDIA","learning_level":"Technical - Intermediate","x_content_types":["Demo"]},{"title":"openfold3 Model by Openfold","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/openfold/openfold3","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"A third-generation biomolecular foundation model that predicts the three-dimensional structures of molecular complexes encompassing proteins, DNA, RNA, and ligands.","document_title":"openfold3 Model by Openfold","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"magpie-tts-multilingual Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/magpie-tts-multilingual","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"The Magpie TTS Multilingual model converts text into audio (speech).","document_title":"magpie-tts-multilingual Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"parakeet-ctc-1.1b-asr Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/parakeet-ctc-1_1b-asr","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Parakeet transcribes audio into text, using spaces and apostrophes where needed.","document_title":"parakeet-ctc-1.1b-asr Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"parakeet-ctc-0.6b-asr Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/parakeet-ctc-0_6b-asr","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Parakeet transcribes audio into text, using spaces and apostrophes where needed.","document_title":"parakeet-ctc-0.6b-asr Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"magpie-tts-zeroshot Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/magpie-tts-zeroshot","technologies":["Nemotron"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"The Magpie TTS Zeroshot model converts text into audio (speech).","document_title":"magpie-tts-zeroshot Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"proteinmpnn Model by IPD","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/ipd/proteinmpnn","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"A deep learning model designed to predict and output the optimal amino acid sequences for a given 3D protein backbone.","document_title":"proteinmpnn Model by IPD","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"Proteina Atomistica Data — Consistency Distilled Synthetic Protein Database","featured":false,"x_formats":["code"],"document_url":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/resources/proteina-atomistica_data?version=release","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"A large-scale dataset containing hundreds of thousands of highly diverse, synthetically generated, fully atomistic protein structures designed to help train AI models for discovering and engineering novel proteins.","document_title":"Proteina Atomistica Data — Consistency Distilled Synthetic Protein Database","learning_level":"Technical - Beginner","x_content_types":["Samples"]},{"title":"CodonFM: A Family of Codon-Resolution Language Models Trained on 130 Million Protein-Coding Sequences","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/CodonFM","technologies":["BioNeMo"],"document_date":"2025-11-01T04:00:00.000Z","short_summary":"CodonFM is an open-source family of codon-resolution language models trained on 130 million protein-coding sequences from over 20,000 species to learn contextual representations for downstream codon-aware tasks.","document_title":"CodonFM: A Family of Codon-Resolution Language Models Trained on 130 Million Protein-Coding Sequences","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Megalodon: Scalable Transformer Models for De Novo 3D Molecular Structure Generation","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/megalodon","technologies":["BioNeMo"],"document_date":"2025-07-22T04:00:00.000Z","short_summary":"Megalodon is a family of scalable transformer models enhanced with equivariant layers and trained using a joint continuous and discrete denoising co-design objective to generate high-quality, realistic de novo 3D molecular structures for drug discovery.","document_title":"Megalodon: Scalable Transformer Models for De Novo 3D Molecular Structure Generation","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"GenMol: A Generative AI Model for Creating Novel Molecules via Masked Discrete Diffusion","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/genmol","technologies":["BioNeMo"],"document_date":"2026-01-14T05:00:00.000Z","short_summary":"GenMol is a generative AI model that utilizes masked discrete diffusion and fragment-based generation to create novel and valid molecules encoded in the SAFE molecular representation for drug discovery.","document_title":"GenMol: A Generative AI Model for Creating Novel Molecules via Masked Discrete Diffusion","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"RNAPro: A State-of-the-Art RNA 3D Folding Model","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/RNAPro","technologies":["BioNeMo"],"document_date":"2026-01-17T05:00:00.000Z","short_summary":"RNAPro is a state-of-the-art RNA 3D folding model developed in collaboration with the hosts and winners of the Stanford RNA 3D Folding Kaggle competition that incorporates RNA-specific modules like template modeling, multiple sequence alignment, and a pretrained RNA language model to enhance structure prediction.","document_title":"RNAPro: A State-of-the-Art RNA 3D Folding Model","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"molmim Model by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/molmim-generate","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Performs controlled generation to discover novel small molecules optimized for specific physical, chemical, or therapeutic properties.","document_title":"molmim Model by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"rfdiffusion Model by IPD","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/ipd/rfdiffusion","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"A generative model of protein backbones used specifically for novel protein scaffolding and protein binder design tasks.","document_title":"rfdiffusion Model by IPD","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"Boltz-2 Model by MIT","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/mit/boltz2","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Predicts complex biomolecular structures and interactions using a generalized architecture for proteins, DNA, RNA, and ligands.","document_title":"Boltz-2 Model by MIT","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"alphafold2 Model by DeepMind","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/deepmind/alphafold2","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Predicts the 3D structure of a protein from its amino acid sequence.","document_title":"alphafold2 Model by DeepMind","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"diffdock Model by MIT","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/mit/diffdock","technologies":["NIM"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Predicts the 3D binding poses and structures of how a small molecule (ligand) interacts with a target protein.","document_title":"diffdock Model by MIT","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"La-Proteina: Partially Latent Flow Matching for Joint Protein Sequence and Atomistic Structure Generation","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/la-proteina","technologies":["BioNeMo"],"document_date":"2025-09-23T04:00:00.000Z","short_summary":"La-Proteina is a generative AI model that uses a partially latent flow matching framework to jointly design fully atomistic protein structures and their corresponding amino acid sequences, enabling the creation of novel proteins and the scaffolding of functional atomic motifs.","document_title":"La-Proteina: Partially Latent Flow Matching for Joint Protein Sequence and Atomistic Structure Generation","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"BioNeMo-MoCo: Modular Co-Design for Generative Model Frameworks","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/bionemo-framework/tree/main/sub-packages/bionemo-moco","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"BioNeMo-MoCo (Modular Co-Design) is a sub-package of the BioNeMo Framework that provides a flexible library of abstracted interpolants for building and sampling from various generative model frameworks, including continuous flow matching, diffusion models (DDPM, VDM), and discrete models (MDLM, D3PM).","document_title":"BioNeMo-MoCo: Modular Co-Design for Generative Model Frameworks","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"BioNeMo-SCDL: Single Cell Data Loading for Foundation Model Training","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/bionemo-framework/tree/main/sub-packages/bionemo-scdl","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"BioNeMo-SCDL (Single Cell Data Loading) is an independent, PyTorch-compatible library that provides high-performance, memory-mapped data loading and a consistent API for scalable training of foundation models using large-scale single-cell genomic datasets.","document_title":"BioNeMo-SCDL: Single Cell Data Loading for Foundation Model Training","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"BioNeMo-Noodles: Memory-Mapped FASTA File I/O for Genomic Sequences","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/bionemo-framework/tree/main/sub-packages/bionemo-noodles","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"BioNeMo-Noodles is a Python wrapper for the noodles library that extends the FAIDX (FASTA Index) format to support memory-mapped file I/O, enabling efficient and scalable access to large-scale FASTA genomic sequence files.","document_title":"BioNeMo-Noodles: Memory-Mapped FASTA File I/O for Genomic Sequences","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"KERMT: A Pretrained Graph Neural Network for Molecular Property Prediction","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/KERMT","technologies":["BioNeMo"],"document_date":"2025-11-19T05:00:00.000Z","short_summary":"KERMT (Kinetic GROVER Multi-Task) is a pretrained graph neural network model designed for high-accuracy molecular property prediction, enhancing the GROVER framework with multitask fine-tuning and GPU acceleration to streamline early-stage drug discovery and ADMET safety testing.","document_title":"KERMT: A Pretrained Graph Neural Network for Molecular Property Prediction","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"nvMolKit: High-Performance GPU-Accelerated Library for Computational Chemistry","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/nvMolKit","technologies":["BioNeMo"],"document_date":"2026-03-09T04:00:00.000Z","short_summary":"nvMolKit is a high-performance, GPU-accelerated cheminformatics library designed to provide massive speedups for core RDKit functions, including molecular similarity calculations, fingerprinting, and 3D conformer generation and relaxation.","document_title":"nvMolKit: High-Performance GPU-Accelerated Library for Computational Chemistry","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"cuik-molmaker: Molecular Featurization for Deep Learning Models","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/cuik-molmaker","technologies":["BioNeMo"],"document_date":"2025-11-05T05:00:00.000Z","short_summary":"cuik-molmaker is a specialized C++ and Python package designed to accelerate molecular featurization by converting chemical structures into optimized formats for deep learning models, such as graph neural networks, while significantly reducing memory usage.","document_title":"cuik-molmaker: Molecular Featurization for Deep Learning Models","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Proteina: A Large-Scale Flow-Based Protein Backbone Generator","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/proteina","technologies":["BioNeMo"],"document_date":"2025-07-24T04:00:00.000Z","short_summary":"Proteina is a large-scale, flow-based generative model for de novo protein backbone design that leverages a tailored scalable transformer architecture and hierarchical fold class labels to generate diverse, designable protein structures up to 800 residues in length.","document_title":"Proteina: A Large-Scale Flow-Based Protein Backbone Generator","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"AvgFlow: Efficient Molecular Conformer Generation with SO(3) Averaged Flow-Matching and Reflow","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/avgflow","technologies":["BioNeMo"],"document_date":"2026-01-08T05:00:00.000Z","short_summary":"avgflow is a framework for fast and accurate molecular structure generation using an SO(3)-averaged flow-matching approach.","document_title":"AvgFlow: Efficient Molecular Conformer Generation with SO(3) Averaged Flow-Matching and Reflow","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"ReaSyn: Synthesis Pathway Prediction from Building Blocks to Final Product","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA-Digital-Bio/ReaSyn","technologies":["BioNeMo"],"document_date":"2026-01-11T05:00:00.000Z","short_summary":"ReaSyn predicts synthesis pathways — the sequence of reaction steps from building blocks to final products — for a target molecule.","document_title":"ReaSyn: Synthesis Pathway Prediction from Building Blocks to Final Product","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"cuEquivariance: Low-Level Primitives and Tensor Ops for Equivariant Neural Networks","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/cuEquivariance","technologies":["BioNeMo"],"document_date":"2026-02-17T05:00:00.000Z","short_summary":"cuEquivariance is an NVIDIA CUDA-X library designed to accelerate the construction and execution of high-performance geometric neural networks by providing optimized CUDA kernels for equivariant operations like segmented tensor products and triangle attention.","document_title":"cuEquivariance: Low-Level Primitives and Tensor Ops for Equivariant Neural Networks","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"Build A Generative Virtual Screening Pipeline Blueprint by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/generative-virtual-screening-for-drug-discovery","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"This blueprint shows how generative AI and accelerated NIM microservices can design optimized small molecules smarter and faster.","document_title":"Build A Generative Virtual Screening Pipeline Blueprint by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"Build A Generative Protein Binder Design Pipeline Blueprint by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/protein-binder-design-for-drug-discovery","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"This blueprint shows how generative AI and accelerated NIM microservices can design protein binders smarter and faster.","document_title":"Build A Generative Protein Binder Design Pipeline Blueprint by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"Parabricks Umi_Fgbio","featured":false,"x_formats":["code"],"document_url":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/containers/clara-parabricks-umi-fgbio?version=1.4.0-1","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Developed by Fulcrum Genomics, utilize molecular tagging (UMIs) to distinguish actual biological variants from machine sequencing errors, ensuring high-fidelity genomic data analysis.","document_title":"Parabricks Umi_Fgbio","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"BioNeMo Recipes","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/bionemo-framework/tree/main/bionemo-recipes","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"BioNeMo Recipes is a collection of reference implementations and step-by-step guides that demonstrate how to efficiently scale and optimize the training of biological foundation models using NVIDIA Transformer Engine and popular open-source frameworks.","document_title":"BioNeMo Recipes","learning_level":"Technical - Intermediate","x_content_types":["Tutorial"]},{"title":"fq2bam (BWA-MEM + GATK)","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_fq2bam.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Part of Burrows-Wheeler Aligner (BWA) bioinformatics software package to provide fast, accurate algorithm for mapping DNA sequence reads to a reference genome.","document_title":"fq2bam (BWA-MEM + GATK)","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"BioNeMo Framework","featured":false,"x_formats":["code"],"document_url":"https://github.com/NVIDIA/bionemo-framework","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"BioNeMo Framework is an open-source, scalable Python library built on PyTorch Lightning and NVIDIA NeMo that provides researchers with optimized tools, model architectures, and data loaders for training and deploying large-scale foundation models in biology and chemistry.","document_title":"BioNeMo Framework","learning_level":"Technical - Beginner","x_content_types":["Samples"]},{"title":"deepvariant","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_deepvariant.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Google's deep-learning based variant caller for germline/genetic variants (i.e., inherited diseases like sickle cell disease).","document_title":"deepvariant","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"rna_fq2bam","featured":false,"x_formats":["webpage"],"document_url":"https://docs.nvidia.com/clara/parabricks/latest/documentation/tooldocs/man_rna_fq2bam.html","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Spliced Transcripts Alignment to a Reference (STAR). Software tool for aligning RNA sequencing (RNA-seq) reads to a reference genome.","document_title":"rna_fq2bam","learning_level":"Technical - Beginner","x_content_types":["Documentation"]},{"title":"Genomics Analysis Blueprint by NVIDIA","featured":false,"x_formats":["hands-on"],"document_url":"https://build.nvidia.com/nvidia/genomics-analysis","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Accelerate genomics analysis workflows from alignment to annotation, using NVIDIA Parabricks.","document_title":"Genomics Analysis Blueprint by NVIDIA","learning_level":"Technical - Beginner","x_content_types":["Demo"]},{"title":"DeepSAP","featured":false,"x_formats":["code"],"document_url":"https://catalog.ngc.nvidia.com/orgs/nvidia/teams/clara/containers/clara-parabricks-deepsap?version=v0.0.3","technologies":["BioNeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"Enhance RNA-seq alignment by combining it with Transformer based Splice Junction Scoring (TSJS). TSJS alone improves accuracy when applied with additional tools including STAR.","document_title":"DeepSAP","learning_level":"Technical - Intermediate","x_content_types":["Samples"]},{"title":"NeMo | Build, Monitor, and Optimize AI Agents","featured":false,"x_formats":["webpage"],"document_url":"https://www.nvidia.com/en-us/ai-data-science/products/nemo/","technologies":["NeMo"],"document_date":"2026-03-10T04:00:00.000Z","short_summary":"NeMo™ is a modular, interoperable, enterprise-ready software suite for managing the AI agent lifecycle, and can be used to develop digital health solutions.","document_title":"NeMo | Build, Monitor, and Optimize AI Agents","learning_level":"Technical - Beginner","x_content_types":["Overview"]}]