librewebui/libre-webui repository overview

⁠Libre WebUI

A local-first workspace for chat, private knowledge, artifacts, and isolated model-driven work.

Self-hosted. Provider-flexible. Apache 2.0. No application telemetry.

Libre WebUI connects to local models through Ollama or to providers you choose. It includes document search, interactive artifacts, personas, multi-user access controls, image and speech providers, and task-scoped Work containers for model-driven file and command workflows.

⁠Start with Docker

The repository includes Compose configurations for bundled Ollama, an existing Ollama server, and NVIDIA GPU inference:

git clone https://github.com/libre-webui/libre-webui.git
cd libre-webui
docker compose up -d

Open http://localhost:8080⁠. The first account created on a fresh installation becomes the administrator; further registration remains disabled unless an administrator enables it deliberately.

See the Docker deployment documentation⁠ for persistent secrets, HTTPS, backups, external Ollama, and GPU configuration.

The image includes the same operational command as npm and Homebrew. For example, inspect a running container's mounted data without using internal build paths:

docker exec libre-webui libre-webui recovery-check --json
docker exec libre-webui libre-webui backup --help

⁠Image tags

• latest — current stable image from the production branch
• dev — current development image; may include unfinished or breaking changes
• Semantic versions such as 0.18.0 — immutable release lines
• Commit tags such as sha-667921d — builds tied to an exact source revision

Images are published for Linux AMD64 and ARM64 to both Docker Hub and GitHub Container Registry.

⁠Other installation paths

• npm: npx libre-webui
• GHCR: ghcr.io/libre-webui/libre-webui
• Kubernetes: helm install libre-webui oci://ghcr.io/libre-webui/charts/libre-webui
• Desktop clients: GitHub Releases⁠

⁠Links

• Website⁠
• Documentation⁠
• Source code⁠
• Release notes⁠
• Security policy⁠
• Project charter⁠