Source

nvidia/skills

447 skills · 412.4K combined installs

Skills from this source

#

Skill

Source

8W Activity

Installs

tao-analyze-gaps-vlm-bcq Official Extract false-positive and false-negative gaps from VLM binary-classification-question (BCQ, yes/no) predictions. Use…

tao-train-mask2former Official Mask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with masked attent…

tao-run-on-brev Official Brev managed GPU instances with Docker support. Use when running TAO training, evaluation, or inference on Brev GPU i…

tao-train-metric-learning-recognition Official Metric-learning recognition (ml-recog) for fine-grained visual recognition. Learns embeddings for retrieval-based mat…

tao-train-rtdetr Official RT-DETR (Real-Time DEtection TRansformer) for 2D object detection. Designed for real-time inference with competitive …

tao-train-visual-changenet Official Visual ChangeNet for binary image classification and segmentation in AOI defect detection. Use when training, evaluat…

tao-train-nvdinov2 Official NVDINOv2 for self-supervised visual representation learning. Trains vision transformers via self-distillation (teache…

tao-train-single-step Official Standard single-step train/eval/export workflow for any TAO model. Use when training a TAO model on a dataset without…

tao-generate-referring-expressions Official Four-step image referring-expression pipeline: turns images plus KITTI bounding-box labels into region descriptions, …

tao-run-deft-aoi Official Run the full DEFT AOI improvement loop for NVIDIA TAO VisualChangeNet / ChangeNet PCB inspection models: baseline eva…

tao-train-centerpose Official CenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations for 6-DoF object p…

tao-train-segformer Official SegFormer for semantic segmentation. Lightweight transformer-based architecture with hierarchical feature extraction,…

tao-run-automl-deft-pipeline Official Run the canonical NVIDIA AOI three-phase training pipeline — Phase 1 AutoML baseline (HPO), Phase 2 DEFT loop (RCA → …

tao-train-pointpillars Official PointPillars for 3D object detection from LiDAR point clouds. Encodes point clouds into a pseudo-image via a pillar-b…

tao-train-mask-auto-encoder Official Masked Auto-Encoder (MAE) for self-supervised pretraining and fine-tuning. Masks random patches and reconstructs them…

tao-train-mask-auto-label Official MAL (Mask Auto-Label) for weakly-supervised segmentation. Produces segmentation masks from minimal annotations (point…

tao-train-ocdnet Official OCDNet for scene text detection. Detects arbitrary-oriented text regions in natural images using a differentiable bin…

tao-validate-dataset-format Official Run `tao-daft validate` to check NVIDIA TAO DAFT datasets for structure, schema, and cross-reference errors. Do not u…

tao-train-bevfusion Official BEVFusion for multi-sensor 3D object detection. Fuses LiDAR point clouds and camera images in bird's-eye-view (BEV) s…

tao-train-mask-grounding-dino Official Mask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for open-s…

tao-train-oneformer Official OneFormer for universal image segmentation. Unifies panoptic, instance, and semantic segmentation with a single archi…

tao-train-deformable-detr Official Deformable DETR for 2D object detection. Uses deformable attention for efficient multi-scale feature processing, ligh…

tilegym-monkey-patch-kernels-to-transformers Official Integrate TileGym kernels into Hugging Face `transformers` models by replacing the library's submodule(s) and certain…

tao-mine-aoi-images Official Runs the DEFT embed-then-mine workflow for VCN AOI iterations — embeds the gap-analysis target parquet, embeds a sour…

tao-run-on-slurm Official Remote SLURM GPU cluster execution over SSH with sbatch/srun, Pyxis/Enroot containers, and Lustre-backed results. Use…

tao-train-sparse4d Official Sparse4D for multi-camera temporal 3D object detection and tracking. Uses sparse queries with deformable attention ac…

tao-train-foundation-stereo Official Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D reconstruction…

tao-train-fast-foundation-stereo Official Real-time stereo depth estimation using FastFoundationStereo (FFS), the distilled bp2 commercial variant of Foundatio…

tao-route-visual-changenet-samples Official Routes the weakest VCN samples (output of `tao-analyze-gaps-visual-changenet`) into per-augmentation-module subsets b…

tilegym-improve-cutile-kernel-perf Official Iteratively optimize cuTile kernel performance through systematic profiling, bottleneck analysis, IR comparison, and …

tao-run-platform Official TAO Execution SDK for submitting and monitoring GPU training jobs on supported platforms (Brev, SLURM, local Docker, …

tilegym-cutile-autotuning Official Use when adding, modifying, optimizing, or debugging CuTile autotuning code.

tilegym-converting-cutile-to-triton Official Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cud…

tilegym-converting-cutile-to-julia Official Converts cuTile Python GPU kernels (@ct.kernel) to cuTile.jl Julia equivalents.

cuopt-user-rules Official Base rules for end users calling NVIDIA cuOpt (routing/LP/MILP/QP/install/server). Not for cuOpt internals — use cuop…

cuopt-server-common Official cuOpt REST server — what it does and how requests flow. Domain concepts; no deploy or client code.

cuopt-skill-evolution Official After solving a non-trivial problem, detect generalizable learnings and propose skill updates. Always active — applie…

cuopt-numerical-optimization-api-cli Official LP, MILP, and QP (beta) with cuOpt — CLI only (MPS/LP/QPS files, cuopt_cli). Use when the user is solving LP, MILP, o…

cuopt-numerical-optimization-api-c Official LP, MILP, and QP (beta) with cuOpt — C API only. Use when the user is embedding LP, MILP, or QP in C/C++.

cuopt-routing-formulation Official Vehicle routing (VRP, TSP, PDP) — problem types and data requirements. Domain concepts; no API or interface.

cuopt-multi-objective-exploration Official Trace, complete, and interpret the Pareto frontier across competing objectives using repeated single-objective cuOpt …

jetson-diagnostic Official Read-only Jetson health snapshot for identity, memory, GPU, thermal, power, storage, services, and top processes.

nemoclaw-user-guide Official Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form.

jetson-memory-audit Official Measure Jetson DRAM/NvMap usage and verify before/after memory reclamation with live audit data.

jetson-llm-serve Official Stand up vLLM or SGLang serving on Jetson, using upstream vLLM on Thor and Orin JetPack 7.2+, and NVIDIA-AI-IOT vLLM …

jetson-inference-mem-tune Official Pick the serving stack and per-runtime memory flags (vLLM, SGLang, llama.cpp, TensorRT Edge-LLM) for an LLM/VLM workl…

jetson-optimize-memory Official >- Reclaim DRAM by disabling unused subsystems across MB1 BCT, MB2 BCT, kernel reserved-memory, and SWIOTLB. Use for …

jetson-quick-start Official >- Entry skill for Jetson / IGX BSP customization. Asks one core click-to-select setup questionnaire and passes prefi…

jetson-llm-benchmark Official Benchmark Jetson LLM/VLM serving performance across vLLM, llama.cpp, and Ollama with structured JSON output.

jetson-headless-mode Official Plan and apply safe Jetson headless-mode changes to reclaim GUI and daemon memory.