NVIDIA

Perf Torch Cuda Graph Specialist

Name: Perf Torch Cuda Graph Specialist
Author: NVIDIA

Development community

SOURCE https://github.com/NVIDIA/TensorRT-LLM/blob/main/.claude/agents/perf-torch-cuda-graph-specialist.md

Description

--- name: perf-torch-cuda-graph-specialist description: > Expert in CUDA Graph capture, replay, and optimization for PyTorch. Delegate to this agent for: (1) Analyzing code for CUDA Graph compatibility, (2) Detecting and eliminating host-device synchronizations, (3) Selecting the right CUDA Graph API (torch.compile, make_graphed_callables, TE, MCore CudaGraphManager, FullCudaGraphWrapper, manual), (4) Implementing capture and replay, (5) Full-iteration CUDA Graph capture, (6) Tro

Installation

Installs to ~/.claude/agents/nvidia-tensorrt-llm-perf-torch-cuda-graph-specialist.md

Terminal

mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/NVIDIA/TensorRT-LLM/HEAD/.claude/agents/perf-torch-cuda-graph-specialist.md -o ~/.claude/agents/nvidia-tensorrt-llm-perf-torch-cuda-graph-specialist.md

Restart Claude Code, or start a new session, for it to be picked up.