brontoguana

brontoguana / ktop

Public

Terminal system resource monitor for hybrid LLM workloads

btop-alternative cli gpu-monitoring htop-alternative llm
64
6
100% credibility
Found Feb 11, 2026 at 43 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

ktop is a terminal dashboard that displays real-time charts for CPU, GPU, memory, network, temperatures, processes, and out-of-memory events, tailored for monitoring intensive AI workloads on Linux.

How It Works

1
🔍 Discover ktop

You hear about ktop, a colorful screen that watches your computer's speed, memory, heat, and internet use while running heavy tasks like AI programs.

2
📥 Set it up easily

Visit the project page, download the files, and run the one-line setup to get it ready on your Linux computer.

3
🚀 Launch your monitor

Type ktop in your terminal and instantly see beautiful charts and bars showing all your computer's vital signs updating live.

4
🎨 Choose a theme

Press t to browse dozens of fun color styles and pick one that makes the display pop just how you like it.

5
📊 Watch everything

Glance at CPU power, memory use, GPU activity if you have one, network speeds, temperatures, top programs, and recent crashes.

✅ Run smoothly

Now you effortlessly keep your computer healthy during big jobs, spotting issues early and staying in control.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 43 to 64 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ktop?

Ktop delivers a sleek Python terminal system monitor tailored for hybrid LLM workloads, displaying GPU utilization and memory (NVIDIA), network upload/download speeds, CPU load, RAM/swap, temperatures with hardware thresholds, and recent OOM kills—all in gradient bars and sparkline histories. It tracks top processes by CPU or memory usage, with a responsive UI polling input every 50ms for instant keyboard navigation. Setup is dead simple: git clone the repo, run setup.sh, and it lands in your PATH as a native command on Linux, pulling in psutil and Rich for the visuals.

Why is it gaining traction?

It crushes bland alternatives like htop or glances with 50 live-preview color themes (hit 't' to switch, persists via config), unique OOM kill tracking from journalctl, and LLM-tuned metrics like per-GPU temps and shared memory in process tables. Network sparklines split up/down directions for intuitive bandwidth spotting, while sim mode fakes loads for testing. Developers dig the no-fluff install and snappy feel over clunky terminal system info tools.

Who should use this?

AI engineers firing up local LLMs via Ollama or vLLM on GPU rigs, chasing OOMs during inference spikes. DevOps folks monitoring terminal system info linux on servers blending CPU, GPU, and network for hybrid workloads. Linux power users wanting prettier, themeable terminal system meaning without leaving the shell.

Verdict

Solid niche pick at 30 stars and 1.0% credibility—mature docs and install beat its early stage, but lacks tests or broad hardening. Grab it for LLM tinkering; otherwise, stick to proven monitors.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.