Open WebUI + Ollama Nextcloud ExApp

A Nextcloud External Application (ExApp) that bundles Open WebUI (chat interface) and Ollama (LLM inference) into a single container managed by Nextcloud's AppAPI.

Features

• All-in-one AI Chat - Open WebUI frontend + Ollama backend in a single ExApp
• Beautiful Chat Interface - Modern, intuitive UI for interacting with AI
• Local LLM Inference - Run models locally with Ollama (no external API needed)
• Auto Model Pull - Automatically downloads a default model on first start
• Conversation History - Persistent chat history with search
• Document Upload - RAG capabilities with file uploads
• Ollama API - Expose Ollama API for other apps (e.g., n8n workflows)
• Mobile Friendly - Responsive design works on all devices

Requirements

• Nextcloud 30 or higher
• AppAPI installed and configured
• Docker with a configured Deploy Daemon (HaRP recommended)
• Sufficient RAM (4GB minimum, 8GB+ recommended for larger models)
• GPU support optional but recommended for performance

Installation

Via Nextcloud App Store

1. Install and enable the AppAPI app in Nextcloud
2. Configure a Deploy Daemon
3. Search for "Open WebUI" in the External Apps section
4. Click Install

Manual Installation

# Register the ExApp with Nextcloud
occ app_api:app:register \
    open_webui \
    <your-daemon-name> \
    --info-xml https://raw.githubusercontent.com/ConductionNL/open-webui-nextcloud/main/appinfo/info.xml \
    --force-scopes

Configuration

Environment Variable	Description	Default
OLLAMA_DEFAULT_MODEL	Model to auto-pull on first start	Not set
OLLAMA_NUM_PARALLEL	Number of parallel Ollama requests	1
OLLAMA_KEEP_ALIVE	How long to keep models loaded	5m
OLLAMA_MAX_LOADED_MODELS	Max models loaded simultaneously	1
OLLAMA_FLASH_ATTENTION	Enable flash attention	false
OPENAI_API_BASE_URL	Additional OpenAI-compatible API URL	Not set
OPENAI_API_KEY	API key for OpenAI endpoint	Not set
ENABLE_SIGNUP	Allow user registration	false

Recommended Models

Model	Size	Best For
llama3.2:1b	~1.3 GB	Quick responses, low resource usage
llama3.2:3b	~2.0 GB	Good balance of speed and quality
mistral:7b	~4.1 GB	High quality general purpose
gemma2:9b	~5.4 GB	Strong reasoning and coding

Usage

After installation, Open WebUI appears as a top menu item in Nextcloud.

Ollama API Access

Other ExApps (like n8n) can access the Ollama API through the proxy:

https://your-nextcloud/index.php/apps/app_api/proxy/open_webui/ollama/api/tags
https://your-nextcloud/index.php/apps/app_api/proxy/open_webui/ollama/api/generate
https://your-nextcloud/index.php/apps/app_api/proxy/open_webui/ollama/api/chat

Or connect directly from other containers on the same Docker network:

http://openregister-exapp-openwebui:11434

Development

Building the Docker Image

docker build -t open-webui-exapp:dev .

Running Locally

docker run -it --rm \
    -e APP_ID=open_webui \
    -e APP_SECRET=dev-secret \
    -e APP_HOST=0.0.0.0 \
    -e APP_PORT=23000 \
    -e APP_PERSISTENT_STORAGE=/data \
    -e NEXTCLOUD_URL=http://localhost:8080 \
    -e OLLAMA_DEFAULT_MODEL=llama3.2:1b \
    -p 23000:23000 \
    open-webui-exapp:dev

Testing Endpoints

# Health check
curl http://localhost:23000/heartbeat

# Ollama API (list models)
curl http://localhost:23000/ollama/api/tags

Architecture

┌─────────────────────────────────────────────────┐
│            Nextcloud + AppAPI                    │
└──────────────────┬──────────────────────────────┘
                   │
                   ▼
┌─────────────────────────────────────────────────┐
│     Open WebUI ExApp Container                   │
│  ┌───────────────────────────────────────────┐  │
│  │  FastAPI Wrapper (port 23000)             │  │
│  │  - /heartbeat, /init, /enabled            │  │
│  │  - /ollama/* → Ollama API (port 11434)    │  │
│  │  - /* → Open WebUI (port 8080)            │  │
│  │  - AppAPIAuthMiddleware                   │  │
│  │  - Iframe loader JS                       │  │
│  └──────────┬──────────────┬─────────────────┘  │
│             │              │                     │
│  ┌──────────▼──────┐  ┌───▼─────────────────┐  │
│  │ Ollama (11434)  │  │ Open WebUI (8080)   │  │
│  │ LLM inference   │  │ Chat interface      │  │
│  │ Model storage   │  │ User management     │  │
│  └─────────────────┘  │ Conversation history│  │
│                        └─────────────────────┘  │
│                                                  │
│  /data/                                          │
│  ├── ollama_models/  (downloaded LLM models)     │
│  └── ...             (WebUI data, secrets)       │
└─────────────────────────────────────────────────┘

Integration Stack

For the best experience, install both ExApps:

1. Open WebUI ExApp - AI chat + Ollama inference (this app)
2. n8n ExApp - Workflow automation with AI (can use Ollama from this app)

License

AGPL-3.0 - See LICENSE for details.

Links

• Open WebUI Documentation
• Ollama Documentation
• Nextcloud AppAPI Documentation
• Conduction