# Nvidia-Status

Nvidia-Status is a self-hosted home automation app. NVIDIA GPU dashboard and exporter that surfaces the GDDR6/GDDR6X VRAM and junction temperatures nvidia-smi hides on consumer RTX 30/40 cards, along with detailed per-GPU stats including per-Docker-container GPU usage. It is open source and self-hosted, installable on Docker Hub and Unraid.

Canonical page: https://ourlocalhost.com/app/2143/nvidia-status

## Description

NVIDIA GPU dashboard and exporter that surfaces the GDDR6/GDDR6X VRAM and junction temperatures nvidia-smi hides on consumer RTX 30/40 cards, along with detailed per-GPU stats including per-Docker-container GPU usage. Exports to Prometheus, a JSON API, MQTT (with Home Assistant auto-discovery) and InfluxDB out of the box for live viewing and long-term analysis in Grafana. Reports core/junction/VRAM temperatures and thresholds, clocks, power and limits, utilisation, fan, PCIe gen/width and live bandwidth, throttle reasons, per-process usage mapped to the owning Docker container, and NVENC sessions. Built for Unraid and works on any Linux host; supports multiple GPUs, optional per-GPU power-limit control, and configuration via the built-in Settings page.

## Platform availability

Available on 1 of 10 platforms: Unraid.

## Health score

42/100 (Fair), based on 3 of 4 factors.
- Maintained: 65 (last commit 131 days ago)
- Popular: 10 (2 GitHub stars)
- Easy to install: 35 (1 install platform, install notes, Docker image)

## Links

- Docker Hub: https://hub.docker.com/r/edddeduck/nvidia-status
- GitHub: https://github.com/edddeduck/nvidia-status
- Unraid: https://ca.unraid.net/apps/nvidia-status-0utzmmf0pafann?q=Nvidia-Status

## Common questions

### Does Nvidia Status require an extra Unraid plugin?

Yes, the Unraid install requires the Nvidia-Driver plugin.

### What container permissions are needed for the hidden temperature readings?

The container must be given SYS_RAWIO and access to /dev/mem. The docker run example also uses the NVIDIA runtime with all visible devices and all driver capabilities.

### Will unsupported junction or VRAM temperature readings break the other metrics?

No, on unsupported cards junction and VRAM temperatures simply show blank. Everything else still works.

### Is an optional power limit re-applied after a restart?

Yes, the per-GPU power-limit control has a latch that re-applies the limit after a restart.

### Where can I change MQTT, InfluxDB, and feature toggles?

Use the built-in Settings page to configure MQTT/InfluxDB and toggle features. The available outputs are dashboard, Prometheus, JSON, MQTT/Home Assistant, and InfluxDB.

---
Source: Our Local Host (https://ourlocalhost.com/app/2143/nvidia-status), the self-hosted and open-source app directory.
