Open WebUI: Feature-Rich Frontend for Ollama & OpenAI

Docker Installation

Open WebUI is distributed as a Docker image. Choose the right command based on your Ollama setup:

# Option 1: Ollama running on the same host machine
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

# Option 2: Ollama also in Docker (use bundled image)
docker run -d \
  -p 3000:8080 \
  --gpus all \
  -v ollama:/root/.ollama \
  -v open-webui:/app/backend/data \
  --name open-webui \
  --restart always \
  ghcr.io/open-webui/open-webui:ollama

# Option 3: OpenAI only (no Ollama)
docker run -d \
  -p 3000:8080 \
  -e OPENAI_API_KEY=sk-your-key \
  -v open-webui:/app/backend/data \
  --name open-webui \
  ghcr.io/open-webui/open-webui:main

# Open http://localhost:3000 — first user becomes admin

Connecting to Ollama & Other Backends

Open WebUI auto-discovers Ollama at http://host.docker.internal:11434. You can add additional OpenAI-compatible connections in Admin Panel > Settings > Connections.

# Environment variables for Open WebUI
# (pass with -e flags in docker run)

# Ollama connection
OLLAMA_BASE_URL=http://host.docker.internal:11434

# Add OpenAI as additional backend
OPENAI_API_KEY=sk-your-openai-key
OPENAI_API_BASE_URL=https://api.openai.com/v1

# LM Studio as additional backend
OPENAI_API_BASE_URL=http://host.docker.internal:1234/v1
OPENAI_API_KEY=lm-studio

# Groq (fast cloud inference)
OPENAI_API_KEY=gsk_your-groq-key
OPENAI_API_BASE_URL=https://api.groq.com/openai/v1

# After connecting, models appear in the model selector dropdown
# You can assign custom names, icons, and descriptions to each model

RAG: Document Upload & Knowledge Base

Open WebUI's RAG system lets you upload documents once and reference them in any conversation. It embeds documents using a local model and stores vectors in Chroma.

# RAG setup workflow:

# 1. Set embedding model (Admin Panel > Documents > Embedding Model)
#    Recommended: nomic-embed-text (via Ollama) or all-minilm (built-in)
#    Pull embedding model: ollama pull nomic-embed-text

# 2. Upload documents:
#    Workspace > Knowledge > Upload (PDF, DOCX, TXT, MD, CSV, web URLs)
#    Or use the paperclip icon in any chat for one-time attachment

# 3. Reference in chat:
#    Type # to search your knowledge base documents
#    Matching documents are injected as context automatically

# 4. Configure chunk settings (Admin Panel > Documents):
#    Chunk Size: 1500 characters (default)
#    Chunk Overlap: 100 characters
#    Top K Results: 5 (docs retrieved per query)

# Supported file types: PDF, DOCX, XLSX, PPTX, TXT, MD, RST, HTML, CSV, JSON, XML

Pipe Plugins: Extending Open WebUI

Pipes are Python functions injected into the message pipeline. Install community pipes from the Open WebUI hub or write your own:

Open WebUI vs AnythingLLM