Koios is a self-hosted industrial inferencing platform for deploying and running ML models.
6.8K
Ai-OPs IoT platform for industrial environments. Koios connects directly to OT devices, collects data in real time, and runs on-device AI/ML inference at the edge — all on-premises, so data never has to leave your site.
Full documentation: https://ai-ops.com/docs
Koios is a single monolithic container that bundles the configuration database, time-series database, in-memory cache, web proxy, and all application services. Docker creates any missing named volumes automatically.
docker run -d --name koios --network host \
--mount source=koios_data_postgres,target=/var/lib/postgresql/16/main \
--mount source=koios_data_influxdb,target=/root/.influxdbv2 \
--mount source=koios_media,target=/var/www/koios/media \
--mount source=koios_logs,target=/var/www/koios/logs \
--mount source=koios_certs,target=/var/www/koios/certs \
--mount source=koios_license,target=/var/www/koios/license \
--mount source=koios_secrets,target=/var/www/koios/secrets \
aiopinc/koios:latest
Then open https://<server-ip> and log in:
adminkoios⚠️ Change the default password immediately after first login. Koios also requires a license to operate; on first login it opens an activation wizard. See Licensing Koios.
Run with --network host (recommended) so the container can reach industrial
devices directly on the host network.
| Port | Purpose |
|---|---|
443 | Web UI / API (HTTPS) |
80 | HTTP (redirects to HTTPS) |
Prefer explicit mapping? Replace --network host with -p 443:443 -p 80:80
(you may then need extra -p flags for device protocol ports).
All data lives in seven named volumes that persist across restarts, updates, and re-deployments. Include all seven in your backups.
| Volume | Contents |
|---|---|
koios_data_postgres | Configuration database (devices, tags, models, settings) |
koios_data_influxdb | Time-series history |
koios_media | Uploaded model files |
koios_logs | Service and device logs |
koios_certs | TLS + OPC-UA certificates |
koios_license | Active license |
koios_secrets | Runtime-generated secrets |
| Resource | Minimum | Recommended |
|---|---|---|
| CPU | 2 cores | 4+ cores |
| RAM | 4 GB | 8+ GB |
| Disk | 20 GB free | 50+ GB free |
Host: Ubuntu Server 20.04 / 22.04 / 24.04 LTS · Docker Engine 20.10+ · NTP-synced clock (required — Koios timestamps all data using the host clock).
| Guide | Link |
|---|---|
| System requirements | https://ai-ops.com/docs/installation/system-requirements |
| Run as a systemd service (production) | https://ai-ops.com/docs/installation/running-as-a-service |
| Manual start (dev/testing) | https://ai-ops.com/docs/installation/manually-starting-koios |
| Licensing | https://ai-ops.com/docs/installation/licensing |
| Environment variables | https://ai-ops.com/docs/installation/environment-variables |
| Backup & restore | https://ai-ops.com/docs/installation/backup-restore |
| Updating Koios | https://ai-ops.com/docs/updates/general |
© Ai-OPs, Inc. · Koios is licensed software. Contact your Ai-OPs representative for a license key.
Content type
Image
Digest
sha256:e5eaa4022…
Size
630.1 MB
Last updated
1 day ago
docker pull aiopinc/koios:v1.1.0-rc.4Pulls:
1,138
Last week