Open WebUI
Self-hosted web interface for chatting with local and cloud AI models
In our rankings
Scores on the self-hosted ai chat interfaces criteria
Evaluated Nov 5, 2026- Services to run ×3 5
5/5 means: One container or one package, with storage built in
The documented Docker install is a single container with one data volume, and a pip package is the alternative.
3 quotes - Model providers ×3 4
4/5 means: As 3, plus at least two named cloud providers, such as OpenAI and Anthropic
Ollama, any OpenAI-compatible API, and OpenAI and Anthropic by name; fewer than ten named providers are listed on the pages we captured.
2 quotes - Users and sign-in ×2 5
5/5 means: As 4, plus user groups and either two-factor sign-in or SCIM provisioning
Roles and groups, SSO through OIDC or LDAP, SCIM provisioning and two-factor sign-in.
5 quotes - Chat with your documents ×2 5
5/5 means: As 4, plus a choice of several vector databases and either hybrid search or several document extraction engines
Built-in knowledge bases with 13 vector database options, hybrid search and 8 document extraction engines.
4 quotes
Evidence (16 quotes)
- 1
“an extensible, easy-to-use, self-hosted platform built to run entirely offline. Connect local models through Ollama or use OpenAI-compatible APIs for local and cloud models.”
github.com/open-webui/open-webui · checked - 2
“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”
github.com/open-webui/open-webui · checked - 3
“Standard image with local AI dependencies”
github.com/open-webui/open-webui · checked - 4
“Requires Python 3.12.”
github.com/open-webui/open-webui · checked - 5
“Chat with Ollama, OpenAI, Anthropic, or any OpenAI-compatible provider from a single interface.”
docs.openwebui.com/features · checked - 6
“RBAC Roles, groups, and per-resource permissions”
docs.openwebui.com/features · checked - 7
“SSO/OIDC/LDAP Federated authentication with any identity provider”
docs.openwebui.com/features · checked - 8
“SCIM 2.0 Automated user and group provisioning”
docs.openwebui.com/features · checked - 9
“Two-Factor Sign-In (MFA) Require a code from an authenticator app at every sign-in”
docs.openwebui.com/features · checked - 10
“13 vector databases ChromaDB and PGVector (officially maintained), plus non-core integrations: Qdrant, Milvus, Elasticsearch, and more”
docs.openwebui.com/features · checked - 11
“Hybrid search BM25 + vector search with cross-encoder reranking”
docs.openwebui.com/features · checked - 12
“8 extraction engines Tika, Docling, Azure, Mistral OCR, Datalab Marker, MinerU, PaddleOCR, custom loaders”
docs.openwebui.com/features · checked - 13
“Upload files, build knowledge bases, and let the AI retrieve exactly the information it needs.”
docs.openwebui.com/features · checked - 14
“Includes code under the Open WebUI License, which requires preserving the Open WebUI branding”
github.com/open-webui/open-webui · checked - 15
“Open WebUI is multi-user from day one.”
docs.openwebui.com/features · checked - 16
“Lightweight image using external services for embeddings, document extraction, and voice”
github.com/open-webui/open-webui · checked