syllo/nvtop

GPU & Accelerator process monitoring for AMD, Apple, Huawei, Intel, NVIDIA and Qualcomm

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 51 minutes ago
Added to GitGenius on September 5th, 2026
Created on June 16th, 2017
Open Issues & Pull Requests: 142 (+0)
GitHub issues: Enabled
Number of forks: 429
Total Stargazers: 10,977 (+0)
Total Subscribers: 85 (+0)

Repository Insights (GitGenius)

Median issue/PR response: 17.7 days
Mean response time: 99.9 days
90th percentile: 378.7 days
Tracked items: 104

How this project is maintained

Around half of the issues opened in the past year never receive a reply. 99% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. Only 5% of issues opened in the past year have been closed. Three people close 64% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 73
New in 7 days: 1
Closed in 7 days: 2
Avg open age: 513 days
Stale 30+ days: 66
Stale 90+ days: 54

Recent activity

Opened in 7 days: 1
Closed in 7 days: 2
Comments in 7 days: 0
Events in 7 days: 4

Top labels

  • enhancement (6)

Most active issues this week

Detailed Description

NVTOP is a command-line GPU and accelerator monitoring tool that displays real-time process information in an htop-like interface.

The tool addresses the need to monitor GPU resource usage across multiple vendors and accelerator types. It works by reading GPU metrics and process information through vendor-specific kernel interfaces and drivers, then presenting this data in a terminal-based interface organized by GPU device. Users can view which processes are consuming GPU resources, memory usage, and other performance metrics in a familiar top-like layout.

NVTOP suits developers and system administrators who need to track GPU utilization on Linux systems with AMD, Intel, NVIDIA, Qualcomm Adreno, Apple, Huawei Ascend, or other supported accelerators. It is particularly valuable in multi-GPU environments where monitoring individual device usage is important. The tool includes an interactive setup window accessible via F2 that allows customization of the display, and preferences can be saved for subsequent runs. Support varies by vendor: AMD requires kernel 5.14 or later with the amdgpu driver, Intel uses i915 or xe drivers, and NVIDIA relies on proprietary drivers. Apple support is noted as limited for M1 and M2 chips.

The project maintains active support across a broad range of GPU vendors, regularly adding and updating drivers as new hardware and kernel interfaces become available. Development includes distribution-specific installation paths for major Linux distributions, containerized deployment options, and comprehensive documentation including a manual page. The codebase is written in C and uses ncurses for terminal rendering, keeping the tool lightweight and portable across different Linux environments.