Open WebUI with Ollama
The template pairs Open WebUI with Ollama for CPU-based inference using two configured small models. Interface language and direction depend on the software version and settings.
Open WebUI template details
The template connects Open WebUI to Ollama and provides persistent volumes. Model use, document retrieval, access control, and data handling depend on configuration.
Ollama and Open WebUI template
The marketplace template pairs Open WebUI with Ollama and pulls Llama 3.2 1B and Qwen 2.5 1.5B models on first startup. Model downloads require network access; inference runs on CPU. Provisioning is currently unavailable in the international sandbox.
Chat with locally configured models
The template connects Open WebUI to Ollama over the service network. Data handling depends on the model and integrations configured in the application.
Run selected language models with Ollama
The template attempts to download Llama 3.2 1B and Qwen 2.5 1.5B on first startup and runs inference on CPU without a GPU. Model availability and performance depend on network access and allocated resources.
Ask questions about documents with retrieval-augmented generation (RAG)
Open WebUI includes document and knowledge retrieval features. Results depend on the model, document quality, and configuration.
Language and text-direction settings
Language and right-to-left layout support depend on the Open WebUI version and settings. Check the interface with the languages and fonts your team uses.
Frequently asked questions about Open WebUI and Ollama
Answers to common questions about the template, model execution, document retrieval, and data handling:
Open WebUI with Ollama: template and configuration overview
Self-hosted Open WebUI gives you control over its deployment and model connections. Data handling still depends on the models, providers, logs, and integrations you configure; review these settings before processing confidential information.
What is Open WebUI?
Open WebUI is an open-source web interface for working with language models. It can connect to Ollama and other supported providers; available models and behavior depend on the configuration.
For document retrieval workflows that use a separate vector database, Open WebUI can be connected to a service such as qdrant if configured for your use case. For visual language-model workflows, see flowise ; for notebook-based analysis, see Jupyter cloud platform is available to you.
Cloud hosting compared with local and self-managed deployment
Language models have different CPU, memory, and GPU requirements. The current marketplace template runs Ollama on CPU and configures two small models. Running a different model may require more resources and additional configuration.
The marketplace template defines separate Open WebUI and Ollama containers with persistent volumes. It attempts to download two models on first startup; downloads require network access. Service provisioning is currently unavailable in the international sandbox.
Open WebUI deployment options
Consider setup, database compatibility, access controls, licensing, and maintenance when comparing Metabase deployment options.
| Feature | Deplomo marketplace template | Hosted AI services | Local installation |
|---|---|---|---|
| Data handling and control | Self-hosted interface and model; data handling depends on the selected integrations | Provider data handling varies by service and plan | Local use avoids hosting a shared team service; collaboration depends on configuration |
| Model availability and inference | Models can run locally after download; model acquisition still requires network access | Hosted models require access to their provider; availability and terms vary | Local inference depends on the computer’s hardware and availability |
| Document retrieval (RAG) | Document features are available in Open WebUI; upload limits and vector database integrations depend on configuration | Document handling and privacy depend on the provider and configuration | Local document processing depends on device resources and software settings |
| Pricing and costs | Fixed monthly EUR package per zone, subject to hard resource limits | Hosted provider prices vary by plan and number of users | Local hardware costs vary by device and usage |
| Interface language and text direction | Check language and right-to-left support in the selected version; models and fonts are configured separately | Interface language and direction depend on the provider and application | A self-managed deployment may require manual client and web-server configuration |
The marketplace template pairs Open WebUI with Ollama. Compare resource needs, model requirements, data handling, and package limits before choosing a deployment.
Open WebUI template requirements
Review the template and its setup requirements before deployment.
- Choose Directus from the marketplace The marketplace lists the Open WebUI and Ollama template; provisioning is currently unavailable in the international sandbox.
- Choose package resources: The template configures two small models for CPU inference. Other models may need more resources and must be configured separately.
- Open the service The template defines two containers with persistent volumes. Domain and TLS configuration depend on the deployment environment.
- Model download and chat The template attempts to download Llama 3.2 1B and Qwen 2.5 1.5B at first startup. Downloads require network access; provisioning is currently unavailable in the international sandbox.
Open WebUI and Ollama template summary
The marketplace template pairs Open WebUI with Ollama and configures two small models for CPU inference. Model downloads, resource needs, data handling, and application settings depend on your configuration.
Review the Open WebUI and Ollama template requirements before deployment.
Run Open WebUI in seconds
Automatic installation, instant setup and custom domain support with a free security certificate.