New in NVIDIA Nsight Visual Studio Code Edition 2026.2
General Enhancements
- Added the Warp Info view, showing every warp and lane on the current GPU with their state, PC, and active/divergent masks, with filtering, sorting, and one-click CUDA focus changes.
- Added instruction-level stepping: Step Over and Step Into advance one GPU instruction at a time when invoked from the Disassembly view.
- Various bug fixes for breakpoints, DEBUG CONSOLE and launch.json configurations.
Intended for Use With
- CUDA Debugger
- See cuda-gdb release notes in the NVIDIA CUDA Toolkit 13.5 (and later).
- IDE Host
- Microsoft Visual Studio Code 1.101 (and later)
- VSCodium / Cursor: current stable versions
- Toolkit
- Workstation/L4T : NVIDIA CUDA Toolkit 13.5 (and later)
- DriveOS (QNX) : NVIDIA DriveOS 7.2.6 (and later)
- Driver
- NVIDIA 621.10 (and later)
Want to know more?
Get to know Nsight VSCE
- Visit the Nsight VSCode Edition overview page
The Nsight VSCode Edition overview page is your information hub for general information, availability, videos, and other links to other NVIDIA tools for GPU code development. - See the Nsight VSCode Edition spotlight video
This Nsight VSCode Edition spotlight shows you how Nsight VSCode Edition fits in with the other NVIDIA IDE debuggers and can be set up in Microsoft’s Visual Studio Code. Then, you’ll see all the key features in action. You’re going to love it! - See Nsight VSCode Edition demonstrated at GTC’21
GTC’21 Video On Demand: Latest Enhancements to CUDA Debugger IDEs - View the Microsoft announcement for the VSCode extension by Nsight VSCE
CUDA Support in Visual Studio Code with Julia Reid - Read the blog postings
Support
- Have a feature request or enhancement suggestion?
Provide feedback. Note that we may not get back to you on these, but will review these entries periodically as we plan our next releases. - Have Questions?
Visit the Nsight VSCE Forum . Our experts and the Nsight VSCE community will be there for you.
Requirements
- CUDA Toolkit : Install the CUDA Toolkit to get important tools for CUDA application development including the NVCC compiler driver and cuda-gdb , the NVIDIA tool for debugging CUDA.
- Microsoft vscode-cpptools : Install Microsoft’s C/C++ for Visual Studio Code to get Intellisense support for CUDA C++ code. Nsight VS Code Edition will automatically install this extension.
Quick Start
Open or create a new CUDA application.
Create a launch configuration .
Create launch configuration833×604 256 KB
Start debugging!