Showing 1231–1260 of 15647 skills
Validate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDeviceOptimizer.
NVIDIA/skills
Validate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.
NVIDIA/skills
Techniques for reducing peak GPU memory in Megatron Bridge — expandable segments, PEFT + SP input re-gather, parallelism resizing, activation recompute, CPU offloading constraints, and common OOM fixes.
NVIDIA/skills
Representative, point-in-time MoE training playbooks by hardware and model family. Use them as candidate seeds, then revalidate the exact runtime, semantics, topology, and steady-state throughput.
NVIDIA/skills
Long-context MoE training guidance for Megatron Bridge. Covers CP sizing, selective recompute, dispatcher choices, and practical patterns from DSV3, Qwen3, and Qwen3-Next long-context experiments.
NVIDIA/skills
Practical guidance for training MoE VLMs in Megatron Bridge. Compares FSDP and 3D-parallel approaches, using rounded lessons from Qwen3-VL, Qwen3-Next, and other multimodal experiments.
NVIDIA/skills
Operational guide for choosing and combining parallelism strategies in Megatron Bridge, including sizing rules, hardware topology mapping, and combined parallelism configuration.
NVIDIA/skills
Validate and use packed sequences and long-context training in Megatron-Bridge, including offline LLM packing, collate-time VLM packing, Energon online packing, and CP constraints.
NVIDIA/skills
Resiliency features in Megatron Bridge including fault tolerance, straggler detection, in-process restart, preemption, and re-run state machine.
NVIDIA/skills
Build and change NVIDIA ALCHEMI Toolkit (nvalchemi) workflows for atomistic simulation and machine-learned interatomic potentials (MLIPs). Use for batched structure relaxation, molecular dynamics, reaction paths, training or fine-tuning an MLIP, atomistic datasets in Zarr, generative structure workflows, model wrapping, and custom hooks or integrators, with the ALCHEMI Toolkit (nvalchemi), or when the user asks for batched GPU atomistic simulation and nvalchemi is installed or acceptable. Not for other simulation packages the user asked for (ASE-only, OpenMM, LAMMPS, GROMACS) when the request does not involve nvalchemi, ALCHEMI Toolkit-Ops (nvalchemiops) kernel work, or plain PyTorch questions.
NVIDIA/skills
Collect and package NVSHMEM put/get bandwidth, latency, and other perftest results with system and topology evidence for performance sanity checks.
NVIDIA/skills
Recommend NVSHMEM NIC-to-PE mappings and environment exports. Use for HCA selection, multi-NIC configuration, or topology-based mapping diagnostics.
NVIDIA/skills
Select an NVSHMEM remote transport from target system and kernel evidence. Use for inter-node selection, compatibility checks, or configuration.
NVIDIA/skills
Diagnose NVSHMEM runtime failures and prepare bug reports for launch, crashes, hangs, correctness, transport, or topology issues.
NVIDIA/skills
Route NVSHMEM tuning to data collection, remote transport, NIC-to-PE mapping, or TMA. Do not use for unrelated CUDA, NCCL, or application tuning.
NVIDIA/skills
Prepare and run PAIDF Cosmos Predict video generation for DEFT media samples.
NVIDIA/skills
Use when a user asks to build, optimize, backtest, rebalance, or analyze a stock portfolio with Mean-CVaR, Mean-Variance/SOCP variance caps, efficient frontiers, scenario generation, or NVIDIA cuOpt.
NVIDIA/skills
Filesystem RAG benchmarks: corpus/, train.json, evaluate_rag.py (RAGAS quality). Not for prod monitoring, latency/throughput benchmarking (use rag-perf), or evals outside this repo layout.
NVIDIA/skills
Point-in-time Run:ai cluster and node-health check: a HEALTHY, DEGRADED, or STUCK verdict per node pool with the dominant blocker — NotReady nodes, offline or failing GPU capacity, GPU counts, allocation vs utilization, the pending backlog, and a stuck scheduler vs a full pool. Use for "is the cluster healthy?", "is the scheduler stuck?", "any nodes down?", "are GPUs failing?", "how many GPUs are free?", "is the cluster saturated?", or "why isn't node pool X scheduling?". Not for quota fairness, team GPU waste, one workload, failed-workload node correlation, XID errors on one node, or writes.
NVIDIA/skills
GPU efficiency check for a Run:ai cluster: are the GPUs that projects, departments, and node pools hold actually computing? Classifies holders as WELL_UTILIZED, UNDER_UTILIZED, or IDLE from average and p95 GPU utilization over one window of any length (default 7 days), quantifies idle GPUs and GPU-hours, and names idle workloads. Use for "which projects hold GPUs but aren't using them?", "GPU utilization by project or department", "where are we wasting GPU cycles?", "is llm-serving actually busy? its GPU memory is full", "is project X actually using its GPUs?", "which workloads in project X are idle?", "which node pool wastes the most GPUs?", or "idle GPU-hours per project last quarter". For "find the idle GPUs and stop them" it answers the read part only. Not for quota fairness, cluster health or free capacity, one workload's status, comparing windows, forecasts, or prices.
NVIDIA/skills
Run `tao-daft convert` to convert NVIDIA TAO DAFT datasets between supported formats. Do not use for non-DAFT data. Use when the user asks to convert a DAFT dataset, change DAFT format, change a TAO dataset format, or run `tao-daft convert`.
NVIDIA/skills
Validate a user AnomalyGenNext dataset and recipe, freeze its validation set, and prepare a canonical texture fine-tuning recipe. Use when AnomalyGenNext task weights do not yet exist. Do not use for ordinary defect generation.
NVIDIA/skills
Cosmos-Embed1 video-text embedding for text-to-video retrieval, video-to-video search, semantic deduplication, and fine-tuning. Use when the user asks to "fine-tune Cosmos-Embed1", "run cosmos-embed inference", "export Cosmos-Embed1", "embed videos", or "search videos with text".
NVIDIA/skills
Kumo-Anomaly — diffusion-based anomaly detection and fine-tuning for multivariate time series. Use when the user asks to "fine-tune Kumo-Anomaly", "run Kumo-Anomaly inference", "detect anomalies with diffusion", "time series anomaly detection", "finetune Kumo-Anomaly", "use perform_anomaly_analysis_with_diffusion", "automl Kumo-Anomaly", "hyperparameter search Kumo-Anomaly", "hyperparameter optimization" or mentions "curriculum_medium.yaml", "final_model.pth", "kumo-anomaly", "sdk/anomaly_analysis", or "TSDiffuser_Generic".
NVIDIA/skills
Kumo-Forecast — transformer-based multivariate time series forecasting with DARR (context-enhanced kNN retrieval), interpretability, and fine-tuning. Use when the user asks to "forecast with Kumo-Forecast", "run forecasting inference", "use perform_forecasting", "DARR mode", "context-enhanced forecasting", "lag horizon attribution", "interpretability", "fine-tune forecasting", "fine-tune forecasting with automl", "hyper-parameter optimization with forecasting", or mentions "kumo-forecast", "moment_head_512_6hr", or "run8_best_model_cr".
NVIDIA/skills
Run AnomalyGenNext 1.1 inference from a native testcase or a prepared generation plan, then publish validated fine-grained and binary COCO labels. Use for synthetic object-detection defect generation, not model training.
NVIDIA/skills
Prepare normalized object-detection false negatives, source masks, and clean-image embedding inputs for AnomalyGenNext. Use when turning box-level FN gaps into a frozen, pair-preserving preparation plan. Do not use for training or generation.
NVIDIA/skills
Routes the weakest VCN samples (output of `tao-analyze-gaps-visual-changenet`) into per-augmentation-module subsets based on each module's label eligibility. Use when the user asks to "route VCN gap samples", "split AOI gaps for k-NN mining and AnomalyGen", or prepare the immediate next step after DEFT gap analysis in a VCN AOI SDA iteration.
NVIDIA/skills
Action recognition from video sequences. Supports RGB, optical flow, and joint (multi-stream) input types for classifying temporal actions in video clips. Use when training, evaluating, exporting, or running inference on a TAO action-recognition model. Trigger phrases include "train action recognition", "video action classification", "RGB + optical flow action model", "TAO ActionRecognition".
NVIDIA/skills
BEVFusion for multi-sensor 3D object detection. Fuses LiDAR point clouds and camera images in bird's-eye-view (BEV) space, used in autonomous driving for robust 3D perception. Use when training, evaluating, or running inference for a TAO BEVFusion model. Trigger phrases include "train BEVFusion", "LiDAR + camera fusion", "BEV 3D detection", "multi-sensor 3D perception".
NVIDIA/skills