Menú
SkillsMP ha recopilado 36 skills de NVIDIA/TensorRT-LLM. Abre una skill para revisar su origen y sus detalles.
Mostrando 36 de 36 skills recopiladas.
Casebook of past successful and classic TensorRT-LLM optimizations (runtime/execution and kernel level) recorded as reusable decision precedents. Consult when deciding which optimization to apply for a classified bottleneck or a given config/model/hardware,…
Idioma del texto original: inglés
Review, design, and refactor TensorRT-LLM PyTorch MoE code for architecture fit, clean code, maintainability, and testability. Always use for any modification, review, refactor, or design planning that touches MoE modules, including…
Idioma del texto original: inglés
Write and implement GPU kernels using NVIDIA CuTe DSL (CUTLASS 4.x Python API) — NOT for Triton, CUDA C++, or conceptual explanations. Trigger only when the user wants to write or implement a kernel, not when asking questions about CuTe DSL concepts or…
Idioma del texto original: inglés
Check the local execution environment for GPU availability, Docker support, and Slurm access. Returns the execution scenario (`satisfied, local, docker`, `satisfied, local, direct`, `satisfied, slurm, local`, or `not_satisfied`), the number of available GPUs,…
Idioma del texto original: inglés
Execute a TensorRT-LLM workload locally in Docker. Runs a fully-resolved Docker command in background, monitors completion, reads logs, and reports results. Workflow-agnostic — does not need to know if the workload is pytest, eval, benchmark, or a custom…
Idioma del texto original: inglés
Submit and monitor a Slurm job on a local cluster. Supports two modes: (1) Persistent allocation (default) — allocates nodes once via nohup salloc, imports the container once, installs once, and reuses across runs by setting SLURM env vars and running the…
Idioma del texto original: inglés
Remote SLURM cluster development via SSH. Use when running jobs, profiling, or developing on a remote SLURM cluster with pyxis/enroot containers. Covers SSH connection management, srun/sbatch/salloc job patterns, tmux-based allocation persistence, file…
Idioma del texto original: inglés
Compile TensorRT-LLM on a SLURM cluster. Covers submitting a batch job with a container image, monitoring the job, and verifying the build. Use when the user wants to compile TRT-LLM remotely via SLURM rather than on a local compute node.
Idioma del texto original: inglés
Performance analysis coordination workflow. Guides profiling delegation, bottleneck classification (compute/memory/launch/communication/sync), and structured report generation. Use when the user asks to analyze performance, profile a workload, check MFU/SOL,…
Idioma del texto original: inglés
Performance optimization coordination playbook. Contains specialist routing table, TileIR two-step pipeline, kernel generation specialist selection, prioritization criteria, and safe modification workflow. Use when the user asks to apply optimizations, write…
Idioma del texto original: inglés
Run TensorRT-LLM test cases, benchmarks, evaluations, or custom scripts by checking the environment (local GPU or Slurm), selecting the appropriate Docker image, and executing either locally or via Slurm job submission. Accepts pre-built command strings —…
Idioma del texto original: inglés
Build Slurm scripts or Docker commands for TensorRT-LLM workloads. Resolves all parameters (docker image, mounts, parallelism, MPI mode), generates the complete script from Category templates, and writes both the script and a job_spec.json manifest to the…
Idioma del texto original: inglés
Runs model-level and module-level tests for TensorRT-LLM. First classifies the test scope (module test or model test), then dispatches to the appropriate workflow. Model tests are further classified by type (functionality/smoke test, benchmark, or…
Idioma del texto original: inglés
Onboard a HuggingFace multimodal model (vision/audio/video + text) to the TensorRT-LLM PyTorch backend. Use when writing a new `tensorrt_llm/_torch/models/modeling_<vlm>.py` plus its input processor and weight mapper, or extending an existing VLM. Not for…
Idioma del texto original: inglés
Compile TensorRT-LLM on a compute node inside a Docker container. Use this when already on a compute node with GPUs visible.
Idioma del texto original: inglés
Adds sharding-aware IR hints (op substitutions, sharding kwargs, all_reduce insertions) directly into an existing AutoDeploy custom model (modeling_*.py). Edits the file in place — no separate _ir.py copy. Validates with apply_sharding_hints and end-to-end…
Idioma del texto original: inglés
Translates a HuggingFace model into a prefill-only AutoDeploy custom model using reference custom ops, validates with hierarchical equivalence tests.
Idioma del texto original: inglés
Debug AutoDeploy accuracy regressions vs a reference score (PyTorch backend or published baseline). Use when an AutoDeploy model's eval score is significantly below the reference and the root cause is unknown.
Idioma del texto original: inglés
Claude Code skill (trtllm-agent-toolkit): implement or extend TensorRT-LLM AutoDeploy fusion transforms under transform/library/ in a TensorRT-LLM checkout. Prefer existing kernels and custom ops; use Triton only when no viable existing-kernel path exists.…
Idioma del texto original: inglés
Enable and interpret TensorRT-LLM AutoDeploy FX graph text dumps via AD_DUMP_GRAPHS_DIR. Use when you need before/after graphs per transform, to locate subgraphs, or to confirm a rewrite ran. Paths and behavior are grounded in tensorrt_llm/_torch/auto_deploy…
Idioma del texto original: inglés
Visualize a specific transformer decoder layer from an AutoDeploy FX graph text dump as a hierarchical DOT/PNG diagram. Optionally annotate nodes with actual GPU kernel names and durations from an nsys trace. Use when the user wants to visualize, inspect, or…
Idioma del texto original: inglés
Optimize existing Triton kernels for NVIDIA TileIR backend on Blackwell GPUs (sm_100+). Adds TileIR-specific autotune configs: occupancy, num_ctas, TMA descriptors. Covers kernel classification (dot-related, norm-like, elementwise, reduction), type-specific…
Idioma del texto original: inglés
ONLY for OpenAI Triton (@triton.jit) kernel development. NEVER use for CUDA C++ kernels, TileIR, or profiling tools (ncu, nsys). The user's request must involve Triton explicitly. Covers Triton-specific patterns: fused elementwise, reductions (softmax,…
Idioma del texto original: inglés
Analyze host/CPU overhead in TensorRT-LLM inference from nsys traces. Detect whether host overhead is the bottleneck using GPU idle ratio, host prep exposed ratio, and per-phase evidence. For regressions, isolate forward steps via allreduce/NVTX patterns,…
Idioma del texto original: inglés
Profiles and optimizes TensorRT-LLM host/CPU overhead using line_profiler (with nsys support planned). Runs iterative profile-analyze-optimize-validate rounds. Use when GPU utilization is low or optimizing PyExecutor throughput.
Idioma del texto original: inglés
Apply CUDA Graphs to PyTorch workloads — API selection (torch.compile, PyTorch make_graphed_callables, TE make_graphed_callables, MCore CudaGraphManager, FullCudaGraphWrapper, manual torch.cuda.graph), code compatibility, capture workflows, dynamic pattern…
Idioma del texto original: inglés
Identify and eliminate host-device synchronizations in PyTorch code. Detects sync points (.item(), .cpu(), boolean indexing, torch.tensor on CUDA), classifies false vs true dependencies, provides sync-free alternatives. Triggers: sync-free, synchronization,…
Idioma del texto original: inglés
Upgrade flashinfer-python version in TensorRT-LLM. Fetches the latest releases from GitHub (stable and nightly), compares with the current pinned version, lets the user pick a target version, and updates all version references across the repo. Use when the…
Idioma del texto original: inglés
Generate a source-backed starting `trtllm-serve --config` YAML for basic aggregate single-node PyTorch serving, aligned with checked-in TensorRT-LLM configs and deployment docs. Preserves explicit latency / balanced / throughput objectives. Excludes…
Idioma del texto original: inglés
Check whether AutoDeploy YAML configs were actually applied by analyzing server logs and optionally graph dumps (AD_DUMP_GRAPHS_DIR). Use when the user wants to verify config application, debug config issues, or check if AutoDeploy transforms (piecewise CUDA…
Idioma del texto original: inglés
Translates a HuggingFace model into a prefill-only AutoDeploy custom model using reference custom ops, validates with hierarchical equivalence tests.
Idioma del texto original: inglés
Analyze ncu (NVIDIA Nsight Compute) profiling output: SOL% bottleneck classification, roofline analysis, occupancy diagnosis, memory hierarchy analysis, warp stall analysis, metric interpretation, and programmatic .ncu-rep report analysis. NOT for kernel…
Idioma del texto original: inglés
Nsight Systems (nsys) CLI for system-level timeline profiling. Use when the user wants to run nsys profile, analyze .nsys-rep reports, use nsys stats/analyze/recipe commands, diagnose GPU idle time from timeline traces, or profile distributed training with…
Idioma del texto original: inglés
Code instrumentation for timing workloads. Two scenarios: (1) Training loop — inject manual timing to report per-iteration latency, throughput (samples/sec), and data load time. (2) Standalone kernel/op — write CUDA event timing code with warmup,…
Idioma del texto original: inglés
Best practices for contributing code to TensorRT-LLM. Covers the official contribution process (issue tracking, fork workflow, DCO signing), coding guidelines, implementation workflow, common mistakes, testing strategy, commit hygiene, and review readiness.…
Idioma del texto original: inglés
Systematic approach to exploring the TensorRT-LLM codebase before implementing new features or optimizations. Teaches how to discover existing infrastructure, trace code paths, and avoid reimplementing what already exists. Derived from real mistakes where…
Idioma del texto original: inglés