Open WebUI — a feature-rich, self-hosted ChatGPT-style web interface for local LLMs (Ollama) and OpenAI-compatible APIs, with chats, RAG, and user management. Deployed as a host-networked Nomad service with a persistent volume; pairs with the ollama pack.
Save as values.hcl, edit, then run:
# The name of the Nomad job.
job_name = "open-webui"
# The Nomad namespace to deploy into.
namespace = "default"
# The datacenters to deploy to.
datacenters = ["*"]
# The Open WebUI container image. Pin a tag in production.
image = "ghcr.io/open-webui/open-webui:main"
# Host port for the Open WebUI web app. Default 3008 to avoid other packs on 3000-3007.
port = 3008
# URL of the Ollama server to use. Defaults to the ollama pack on the same node (host networking).
ollama_base_url = "http://127.0.0.1:11434"
# Secret key for signing sessions (WEBUI_SECRET_KEY). Set a stable random value in production. Empty = Open WebUI generates one and persists it in the data volume.
secret_key = ""
# Docker named volume for /app/backend/data (SQLite DB, chats, users, settings). Open WebUI runs as root, so a fresh volume is writable. Back it up.
data_volume = "open_webui_data"
# Placement constraints — pin to one node so the local volume stays put (ideally the node running ollama). On a nomploy cluster: attribute = "$${meta.nomploy_control_plane}", operator = "=", value = "true".
constraints = []
# The task resources.
resources = {
cpu = 500
memory = 512
}
| Name | Type | Default | Description |
|---|---|---|---|
| job_name | string | "open-webui" | The name of the Nomad job. |
| namespace | string | "default" | The Nomad namespace to deploy into. |
| datacenters | list | ["*"] | The datacenters to deploy to. |
| image | string | "ghcr.io/open-webui/open-webui:main" | The Open WebUI container image. Pin a tag in production. |
| port | number | 3008 | Host port for the Open WebUI web app. Default 3008 to avoid other packs on 3000-3007. |
| ollama_base_url | string | "http://127.0.0.1:11434" | URL of the Ollama server to use. Defaults to the ollama pack on the same node (host networking). |
| secret_key key | string | "" | Secret key for signing sessions (WEBUI_SECRET_KEY). Set a stable random value in production. Empty = Open WebUI generates one and persists it in the data volume. |
| data_volume | string | "open_webui_data" | Docker named volume for /app/backend/data (SQLite DB, chats, users, settings). Open WebUI runs as root, so a fresh volume is writable. Back it up. |
| constraints | list | [] | Placement constraints — pin to one node so the local volume stays put (ideally the node running ollama). On a nomploy cluster: attribute = "$${meta.nomploy_control_plane}", operator = "=", value = "true". |
| resources | object | {
cpu = 500
memory = 512
} | The task resources. |
No variables match.
Open WebUI — a feature-rich, self-hosted ChatGPT-style interface
for local LLMs via Ollama (and any OpenAI-compatible API): multi-model chats, RAG over
your documents, prompts, and user management. Host-networked Nomad service with a persistent
volume. Pairs with the ollama pack.
nomad-pack registry add nomploy github.com/Nomploy/nomad-packs
nomad-pack run open-webui --registry nomploy
In nomploy: create a Compose service, type Nomad Pack, pack open-webui, custom registry
github.com/Nomploy/nomad-packs, then Deploy. Open http://<node-ip>:3008 and create the
admin account (the first user is the admin).
Run the ollama pack alongside it (and ollama pull <model>), so there's a model to chat
with. By default Open WebUI talks to Ollama at http://127.0.0.1:11434 (same node).
| Variable | Default | Notes |
|---|---|---|
image |
ghcr.io/open-webui/open-webui:main |
Pin a tag in production. |
port |
3008 |
Web app. |
ollama_base_url |
http://127.0.0.1:11434 |
The ollama pack on the same node. |
secret_key |
"" |
Set a stable random value in production. |
data_volume |
open_webui_data |
Chats/users/settings. Back it up. |
constraints |
[] |
Pin to a node (ideally the one running ollama). |
resources |
cpu 500 / mem 512 |
Raise for heavy use / RAG. |
count is fixed to 1 (local volume). Pin with constraints.