
I released a local dashboard for Ollama that lets you monitor and control your local AI setup
I’ve been building a side project for people running AI models locally with Ollama, and I just released it publicly.
It’s called OpenSourcesAI Command Center.
The goal is simple: give you one local dashboard where you can see what Ollama and your computer are actually doing.
It can show you:
- Your GPU, VRAM, RAM, CPU, temperature, power use, and other live stats
- Which Ollama models are installed and currently loaded
- Whether a model is running fully on your GPU or falling back to the CPU
- Which models should fit your hardware
- Controls to load or unload installed models
- Local AI chat with speed measurements for each response
- Performance benchmark results
- Local MCP tools and their status
One thing I learned while building this is that hardware detection is surprisingly messy. Different parts of Windows can even report different amounts of VRAM.
Instead of hiding those disagreements, Command Center shows them so you can see what your system is actually reporting.
Everything also stays local. The dashboard runs on 127.0.0.1 and communicates only with AI runtimes running on your own machine.
I just released v0.2.0, including the new interface shown in the screenshot.
Easy to try:
- Open PowerShell or your terminal:
- Paste the code:
​
npx @opensourcesai/cmdcenter serve
Then open the Command Center in your browser:
GitHub Repo:
https://github.com/endoftheline818/opensourcesai-cmdcenter
It’s free, open source, and MIT licensed.
I’d love feedback, especially from people testing it on different hardware. AMD, Intel, Apple Silicon, NVIDIA, multi-GPU systems, and home servers are all useful test cases.