This page's audience our own tag's counts — open to see them

Period

–Visits

–Page views

–Clicks to vendors

–Time on page

–Reading now

Clicks to vendors, by tool

Top countries

Devices

– · visits, page views and clicks counted by our own tag, which counts more than our checked analytics; every figure rounded down · how we count

CUDA-GDB

Free#8 of 27 in Debuggers

CUDA-GDB: A capable open-source debugger for inspecting CUDA CPU and GPU execution together. Ranked #8 of 27 in Debuggers by our editors (8.3/10); pricing: Open source; best for CUDA CPU/GPU debugging.

8.3/10Editor score

Visit NVIDIA

CUDA-GDB8.3 Visit NVIDIA

At a glance

Editor score
8.3 / 10
Pricing
Open source
Best for
CUDA CPU/GPU debugging
Paid from
Free
Supported languages
C/C++, Fortran
Founded
1993 · Santa Clara, California, United States
Facts checked
20 Sep 2026

Where it wins
- Debugs CPU and GPU code simultaneously
- Supports CUDA C/C++, Fortran, remote sessions, and core dumps
- Inspects GPU state across multi-GPU debugging sessions
Where it doesn't
- Focused on NVIDIA GPU development
- Listed platform coverage centers on Linux
- Command-line workflow may not suit every development team

Our verdict on CUDA-GDB

CUDA-GDB is NVIDIA's open-source extension to the GNU Project Debugger for debugging CUDA applications on actual NVIDIA hardware. It is aimed at developers working with CUDA C/C++, CUDA Fortran, and mixed CPU/GPU programs who need one debugger for host and device execution. Its core workflow includes source-level debugging, GPU assembly inspection, single-stepping, kernel-entry breakpoints, and conditional breakpoints based on CUDA thread and block variables. Variable, memory, register, and GPU-state inspection add visibility into device behavior that is difficult to diagnose from CPU-side debugging alone.

The published distribution model is straightforward: CUDA-GDB is included with NVIDIA's CUDA Toolkit and is described as open source, rather than presented as a standalone commercial product. Its feature set also covers multi-GPU and multiple-session debugging, remote debugging through cuda-gdbserver, and inspection of GPU and GPU-plus-CPU core dumps. The documentation describes Linux and QNX targets, while the listed platform coverage is Linux. CUDA-GDB also integrates with DDD, Emacs, Nsight Eclipse Edition, and Nsight Visual Studio Code Edition, giving teams several ways to connect the debugger to an existing development workflow.

CUDA-GDB is a strong fit when diagnosing kernels, GPU memory, thread behavior, registers, or mixed host-device execution is central to the work. Its narrow focus is also its main boundary: this is a CUDA debugger for NVIDIA hardware, not a general-purpose debugger for varied accelerator ecosystems. The command-line foundation may appeal to developers who want direct control and scriptable debugging, while teams seeking a broader visual debugging environment may prefer a tool aligned more closely with their existing IDE. Choose CUDA-GDB for detailed CUDA execution analysis, remote debugging, and dump inspection; consider an alternative when wider platform coverage matters more than CUDA-specific depth.

Your product? Our editors can test CUDA-GDB in depth and publish a review, labelled as paid · from $149/mo →

CUDA-GDB pricing

Plans Open sourceFree Free to use — no paid tier required for the core job.

See plans on developer.nvidia.com

CUDA-GDB fact sheet

Free plan	Not verified
Paid from	Free
Supported languages	C/C++, Fortran
Target platforms	Linux, QNX
Remote debugging	Yes
Conditional breakpoints	Yes
Exception breakpoints	Not verified
Time-travel debugging	Not verified
Core dump debugging	Yes
Deployment	Desktop
Platforms	Linux
Support	Email, Community, Docs
Built for	Solo, Small business, Mid-market, Enterprise (editorial estimate)
Integrations	4 integrations: DDD, Emacs, Nsight Eclipse Edition, Nsight Visual Studio Code Edition
Pricing	Open source
Website	developer.nvidia.com
Facts checked	20 Sep 2026

CUDA-GDB integrations

CUDA-GDB lists 4 integrations on its own site.

DDD
Emacs
Nsight Eclipse Edition
Nsight Visual Studio Code Edition

Alternatives to CUDA-GDB

See all CUDA-GDB alternatives →

CUDA-GDB vs the competition

Compare CUDA-GDB with any tool side by side →

Used CUDA-GDB? Be the first to review it

The editor score above is our own research. What this page doesn't have yet is a reader's view — what you used CUDA-GDB for, what worked and what didn't. No stars are seeded and no review is paid for; an editor reads every one before it appears.

Write a reviewTwo minutes · verified accounts only · read by an editor before it appears

Reviews come only from verified accounts. Sign in or create an account first — your e-mail is never shown.

Title

Your review

0 characters · at least 80, up to 3,000

Pros (optional · one per line)

Cons (optional · one per line)

Your name (shown with your review)

Role (optional)

Company (optional)

How long have you used it?

I've used CUDA-GDB myself and this is my honest opinion — nobody paid or asked me to write it.

Posted from your verified account. Reviews appear after an editor reads them, usually within two working days.

Featured on iTechGuides

CUDA-GDB is listed in our Debuggers directory. Add the badge to your site — it links back to this page.

<a href="https://www.itechguides.com/products/cuda-gdb/"><img src="https://www.itechguides.com/best/badge/cuda-gdb.svg" alt="Featured on iTechGuides" width="230" height="46"></a>

Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026

Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.

Last updated 7 Oct 2026 · How we research and update