Skip to main content
Ollama is an open-source platform (Ollama GitHub) for running local large language models (LLMs) such as Llama. Thanks to a thriving community, you can extend Ollama with interfaces like Chatbot UIs, RAG chatbots, and Kubernetes Helm charts. This guide explores these community integrations and walks through setting up Open Web UI, a ChatGPT-like interface for your local models.

​
Community Integration Categories

The Ollama community maintains several integration types to enhance your local LLM workflow:
Integration CategoryPurposeExample Project
Chatbot UIWeb-based chat interface for local modelsOpen Web UI
RAG ChatbotRetrieval-Augmented Generation with PDF/file uploadollama-rag-ui
Helm PackageDeploy Ollama and models via Kubernetes Helm chartsollama-helm-chart
The image shows a slide titled "Community Integrations" with three sections: "Chatbot UI," "RAG Chatbot," and "Helm Package," each with an icon and numbered 01 to 03.
Later in this article, we’ll link to each project. First, let’s focus on Open Web UI, a widely adopted ChatGPT-like interface for any local model managed by Ollama.

​
Installing and Running Open Web UI

Open Web UI runs inside a Docker container and auto-detects your Ollama agent and locally downloaded models. By default, it listens on port 3000, but you can adjust it as needed.
  • Docker installed on your machine
  • Ollama set up with at least one local model
  • Port 3000 available (or choose another port)

​
Launching with Docker

Run the following command to start Open Web UI: 
docker run -d \
  -p 3000:8080 \
  --add-host=host.docker.internal:host-gateway \
  -v open-webui:/app/backend/data \
  --name open-webui \
  --restart always \
  ghcr.io/open-webui/open-webui:main
If the image isn’t present locally, Docker will pull it automatically: 
Unable to find image 'ghcr.io/open-webui/open-webui:main' locally
main: Pulling from open-webui/open-webui
...
Status: Downloaded newer image for ghcr.io/open-webui/open-webui:main
You can now access the UI at http://localhost:3000.
The image shows a search bar with the text "localhost:3000" and a Docker whale icon below it, labeled "Open WebUI."

​
Use Case: Growmore Investment Firm

Consider Growmore, an investment firm that values AI productivity without risking sensitive client data to the cloud. Their non-developer staff prefer a familiar chat interface over a terminal.
The image illustrates "The Story of Growmore," showing a flow from Growmore to chatbots, then to Ollama, and finally to non-developers.
Running models locally with Ollama ensures that sensitive data never leaves your network. Combine this with community UIs for a secure and user-friendly experience.
By pairing Ollama’s local execution with Open Web UI, Growmore achieves both data privacy and a cloud-like interface similar to ChatGPT or Claude.

​
Exploring Local Models

Once Open Web UI is live, it lists all models available to your Ollama agent. You can effortlessly switch between models like llama3.2, llama2-5.3, Qwen, and Mistral.
The image shows a user interface for selecting models in Open WebUI, with a dropdown menu listing various model options like "llama3.2:latest" and "ph3:latest." The interface also includes a workspace and chat suggestions.

​
Interactive Chat Interface

The chat window allows you to converse with any selected model and even change the model mid-conversation. For example, ask llama3.2 one question, then switch to llama2-5.3 for a follow-up.
The image shows a chat interface with a conversation between a user and two AI language models, discussing their capabilities and readiness to assist. The title at the top reads "Using the Open WebUI Chat Interface."
This flexibility streamlines your local LLM workflow, letting you compare responses and choose the best model for each task.

​
Next Steps and References

In upcoming sections, we’ll dive into other community integrations—such as RAG chatbots and Helm charts—and provide step-by-step setup guides.

Watch Video