CLUSTER METRICS-AMD-DISPLAYS · 20 ENTRIES · EARLIEST ENTRY 2026-07-02
GPU and system monitoring dashboards
This cluster comprises 26 tools for real-time hardware monitoring, primarily GPU-centric, covering utilization, temperature, clock speeds, VRAM, power, and process stats across NVIDIA, AMD, and Intel platforms. Dated entries span from March 2019 to July 2026, with sporadic early activity (one or two per year) and a denser, irregular pattern in 2026 (March–July: 1, 3, 2, 1, 5 entries). Traction is led by spark-dash (7 stars), followed by sparktop (6 stars) and vllm-sentinel (3 stars).
Entry timeline
ENTRY TIMELINE
Each bar is the month a member carries, which is a creation date where the source published one and a crawl date otherwise. Nothing in the index was found before 11 August 2026, and coverage broadened sharply on 14–15 August, so a bar from then on reports how much we found that month, not how much was built; an earlier bar reports how far back a member's own source dated it — not how long we have been watching.
Members by activity
Activity signals, not revenue.
Build origin · 20 unlabelled
Operator · 4 hobby · 15 open source · 1 unlabelled
Two labels per entry, never collapsed into one: how it was built, and who stands behind it. Counted over the 20 listed below. Unlabelled means nothing has decided yet — not that the answer is small.
- 01spark-dash8★26 days—hobbyDashboard that monitors GPU health and inference model metrics for a home cluster of AI servers.
- 02icue-link-lcd-reactor2★18 days—hobbyHardware dashboard that shows live language-model server and GPU metrics on a PC cooling pump's LCD screen under Linux.
- 03siltide-bin0★6 daysliveopen sourceTerminal monitor that displays utilization and status of GPUs, NPUs, and AI accelerators for developers.
- 04LLM-Control-v20★20 days—Provides a graphical interface for controlling a local LLM server running on a GPU.
- 05streamdeck-dashboard1★1 mo—hobbySystem dashboard for Elgato Stream Deck that displays CPU, GPU, temperature, and power metrics on Linux.
- 06gpur-bin0★2 moliveopen sourceTerminal dashboard that monitors GPU metrics across NVIDIA, AMD, and Intel hardware.
- 07vitals0★28 days—hobbyReal-time machine utilization dashboard showing CPU, memory, storage, network, and NVIDIA GPU metrics with rolling history.
- 08cxgpu-bin0★2 moliveopen sourceTerminal GPU monitor that displays real-time performance metrics for NVIDIA and AMD graphics cards.
- 09recipe0★2 moliveopen sourceGPU scheduling tool that plans and executes compute workloads across AMD and NVIDIA hardware.
- 10amdtop0★2 moliveopen sourceTerminal system monitor that displays metrics for AMD GPUs, CPUs, and XDNA NPUs.
- 11linsight0★2 moliveopen sourceSystem monitoring dashboard that displays real-time hardware metrics on Linux machines and supports multiple GPUs.
- 12narsil0★3 moliveopen sourceTerminal resource monitor that displays CPU, memory, and GPU usage with braille charts and labeled bars.
- 13zml-smi-bin0★4 moliveopen sourceProvides hardware diagnostics and monitoring for GPUs, TPUs, and NPUs.
- 14danger-monitor0★4 moliveopen sourceSystem and GPU monitoring application that displays performance metrics in high-resolution charts with customizable themes.
- 15nv-monitor0★5 moliveopen sourceCommand-line tool for NVIDIA GPU that shows live performance, logs data, and exports metrics.
- 16gpufetch0★6 moliveopen sourceTerminal app that shows GPU model, temperature, and usage statistics for system monitoring.
- 17amdguid-glow-bin0★1 yrabandonedliveopen sourceAMDGPU monitoring tool that shows GPU temperature and fan speed on Linux.
- 18tegratop0★1 yrabandonedliveopen sourceTerminal monitoring tool that shows system resource usage and hardware metrics on NVIDIA Jetson boards.
- 19led_matrix_monitoring0★1 yrabandonedliveopen sourceDaemon that renders system metrics on an LED matrix connected to a specific laptop model.
- 20python-gpustat0★3 yrabandonedliveopen sourceA command-line utility that reports GPU memory, utilization, and running processes.
Ranked by traction, highest first, and paged 20 at a time — this is a window on the cluster, never the whole of it. Ties are broken by a stable id so a product cannot appear on two pages or fall between them. Browse every cluster.