NVIDIA TAO
Official@nvidia-tao
Offers a comprehensive framework for training, fine-tuning, and deploying specialized computer vision and deep learning models across diverse GPU-accelerated infrastructure.
Agent Skills by NVIDIA TAO
Showing 60 vetted skills indexed across 1 GitHub repositories.
NVIDIA-TAOREPLACE-WITH-PLATFORM-NAME
Automate GPU job deployment on REPLACE-PLATFORM via APIs and Docker.
NVIDIA-TAOreplace-with-skill-name
Convert PDF documents to plain text for editing.
NVIDIA-TAOExtract text and structured data from PDF documents using pypdf and pdfplumber.
NVIDIA-TAOmodel-trainer
Configure environments and execute AI model training workflows.
NVIDIA-TAOskill-skeleton
Generate a template for new data transformation skills with metadata and instructions.
NVIDIA-TAOtao-launch-workflow
Automate TAO workflow launches with platform selection, credentials, and preflight checks.
NVIDIA-TAOtao-list-capabilities
List TAO Skill Bank plugin capabilities for applications, workflows, models, and AutoML.
NVIDIA-TAOtao-run-on-kubernetes
Submit TAO container jobs as Kubernetes Jobs with NVIDIA GPU scheduling.
NVIDIA-TAOtao-run-on-brev
Automate TAO training, evaluation, and inference jobs on Brev GPU instances via the Brev CLI.
NVIDIA-TAOtao-run-platform
Submit and monitor GPU training jobs on NVIDIA TAO platforms.
NVIDIA-TAOtao-run-on-local-docker
Execute TAO SDK jobs in Docker containers with NVIDIA GPU acceleration.
NVIDIA-TAOtao-run-on-docker
Run Docker containers on GPU hosts with NVIDIA Container Toolkit.
NVIDIA-TAOtao-setup-nvidia-gpu-host
Install NVIDIA drivers, CUDA Toolkit, and Container Toolkit for Docker or Kubernetes backends.
NVIDIA-TAOtao-run-on-slurm
Automate remote TAO GPU job submission on SLURM clusters via SSH.
NVIDIA-TAOtao-train-grounding-dino
Automate Grounding DINO training, evaluation, and ONNX export with PyTorch.
NVIDIA-TAOtao-finetune-cosmos-reason
Automate fine-tuning of video QA models with SFT and LoRA.
NVIDIA-TAOtao-train-mask2former
Automates Mask2Former model training, evaluation, export, quantization, and inference.
NVIDIA-TAOtao-train-action-recognition
Train, evaluate, export, and run inference on TAO action-recognition models.
NVIDIA-TAOtao-train-segformer
Automate training, evaluation, export, and inference for NVIDIA TAO SegFormer models.
NVIDIA-TAOtao-train-metric-learning-recognition
Train, evaluate, export, and infer TAO metric-learning recognition models.
NVIDIA-TAOtao-train-centerpose
Automate training, evaluation, and deployment of 6-DoF pose estimation models.
NVIDIA-TAOtao-finetune-cosmos-embed
Fine-tune and evaluate the Cosmos-Embed1 video-text embedding model.
NVIDIA-TAOtao-train-fast-foundation-stereo
Train, evaluate, export, and run inference on TAO FastFoundationStereo models.
NVIDIA-TAOtao-finetune-clip
Fine-tune CLIP models for image-text retrieval, classification, and ONNX/TensorRT deployment.
Frequently Asked Questions About NVIDIA TAO
FAQPage SchemaWhat specific model training tasks are supported by NVIDIA TAO?▼
NVIDIA TAO supports training, evaluation, and export for diverse architectures including DINO, Mask2Former, SegFormer, PointPillars, and Visual ChangeNet. It enables fine-tuning of vision-language models and provides specialized capabilities for pose estimation, action recognition, and scene text detection.
Which infrastructure environments can execute these model training jobs?▼
Jobs can be executed across local Docker containers, Kubernetes clusters, SLURM environments, or cloud-based GPU instances like Brev. The framework handles job submission, scheduling, and monitoring through integrated connectors for these specific compute backends.
What are the prerequisites for running these model training jobs?▼
Users must have NVIDIA drivers, the CUDA Toolkit, and the NVIDIA Container Toolkit installed on their host machines. Additionally, datasets must be validated for format compatibility using the provided dataset conversion and validation utilities before initiating training.