Open-Source Local LLM Monitoring Dashboard
Important_Quote_1180 · reddit · 2026-07-11
The author has open-sourced a single-file, zero-dependency monitoring dashboard for local LLM servers, designed specifically for DIY inference boxes.
Key Features
- The frontend is just a single index.html that can be opened directly via file://
- The backend is a single Python standard library script
- Reads data via nvidia-smi and each service's Prometheus /metrics
What You Can Monitor
- GPU utilization, VRAM, power draw, temperature, clock speeds, and active processes
- llama.cpp and vLLM metrics like tokens/s, request counts, KV/context usage, and LoRA
- Auto-discovery of worker ports
- Optional support for multiple secondary servers
- System metrics like CPU, RAM, load, network, and disk
- Model inventory lists with quantization/context/throughput details
- Supports extracting reasoning/CoT for each request via logs as a "thought tap"
More from Infra
- Vercel AI Gateway data shows Anthropic, OpenAI and Google at 97.09% spend share — cramforce · 2026-07-21
- NVIDIA starts rolling out 102.4 Tbps Spectrum-6 switches for Vera Rubin AI factories — nvidia · 2026-07-21
- Mustafa Suleyman says Microsoft is preparing for an OpenAI exit, while a new chip costs 30% less than GB200 — thoefler · 2026-07-21
- Microsoft and Mistral sign multi-billion-dollar deal to expand AI infrastructure in Europe — The Decoder · 2026-07-21
- Speculative decoding boosts Qwen3.6-27B on one 5090, but slows crowded servers — luke_pacman · 2026-07-21
- NVIDIA says Blackwell Ultra hit 1,648 TFLOPs per GPU on DeepSeek-V3 671B training — NVIDIAAI · 2026-07-21