Local & self-hosted appsTimothy Jaeryang Baek — began Sept 2023 as "Ollama WebUI", renamed Open WebUI in Jan 2024 (Ollama raised a trademark concern); now community-maintained
Open WebUI
Decision matrix
Open WebUI runtime decision matrix
Overview
Open WebUI is a free, self-hosted browser chat interface that connects to local (Ollama) or hosted model endpoints, so your team gets a shared chat UI without each person calling raw APIs. Pick it when you already have a working model connection and need a shared interface on top. Since v0.6.6 the license requires visible branding unless you're 50 users or fewer, a substantive contributor, or buy an enterprise license.
Other good fits
A team of 50 or fewer users that wants to self-host a chat UI and remove the default branding under the license's small-deployment exemption.
+ Pros
Read pros
Open WebUI puts a browser chat interface in front of model endpoints you configure: a message you type gets sent to a connected local or hosted provider, that provider performs the actual inference, and tokens stream back to the browser. The same interface can point at very different back ends (a local Ollama instance, a hosted API) with very different data destinations, so which one is connected matters more than the interface itself.
Pick it when…
You already have a working local (Ollama) or hosted model connection and want a shared browser interface instead of everyone calling raw APIs.
Your team is 50 users or fewer and wants the small-deployment license exemption to remove default branding.
You want persistent chat history, multiple users, and centralized connection management in one place.
− Cons
Read cons
A shared interface can't guarantee the underlying model's quality, speed or availability; those depend entirely on whatever you connected. Provider compatibility is protocol-specific, so not every model endpoint will work without configuration. Because the browser address is local, it's easy for users to assume everything is running locally even when the configured connection is actually a hosted provider; make the actual provider visible to your users rather than letting the local URL imply local inference.
Pick something else when…
You're a single desktop user and don't need a shared, multi-user interface.LM Studio or Jan are simpler for that case.
The consumer of the model is already an application, not a person typing in a browser. Call the API directly and skip running an extra interface.
You want a full team AI workspace with reusable Skills and workflows, not just chat.Langdock covers more than a chat front-end.
Avoid if
You need unrestricted white-labeling above 50 users without an enterprise license or a substantive-contributor exemption.
Requirements
Open WebUI needs a place to run (a container is the documented route) and persistent storage for its own data: chats, users and settings live in the WebUI host's own database, separate from wherever the model itself runs. It doesn't supply any model capacity on its own; you still need a working Ollama installation or a hosted API key configured as a connection.
First run
Documentation-based suggestion — not run here.
Draft a short notice through an existing local model connection.
As an admin, connect a working Ollama endpoint (or another provider) under Settings → Admin → Connections.
As a user, select that model in the chat dropdown.
Ask it to draft a short, fictional event notice with specific, checkable details (day, time, one restriction).
Compare the response against what you asked for, and confirm the model shown in the dropdown is actually the one that answered.
Done when the draft preserves every detail you specified and you've confirmed which model and connection actually produced it, not just assumed.
Integration
Admins add provider connections under Settings → Admin → Connections, entering the endpoint and its key there rather than in chat messages. Keep a stable secret key and a pinned (not rolling) image tag across container recreation, since the quick start specifically warns about both; some slim container images download model-related components later rather than bundling everything up front, so a small initial image isn't proof the setup is fully offline.
Check the result
If no model appears in the dropdown, check the connection and provider availability in Admin settings; reloading the page repeatedly won't install a model. If a provider rejects a request, resolve that connection before adding a second paid provider just to get an answer.
Links & evidence
Documentation-based · not hands-on · Sep 23, 2026
Rests on Open WebUI's quick start, provider-connection guide and license explanation, checked 23 September 2026. No installation, provider connection or browser session was performed for this review.
Open WebUI quick start · Sep 23, 2026 — Persistent chats/settings storage, stable secret key, rolling vs. pinned image tags.
License explanation · Sep 23, 2026 — Branding requirement since v0.6.6, 50-user exemption, substantive-contributor and enterprise-license exceptions, BSD-3-Clause legacy code through v0.6.5.
Open WebUI · Sep 23, 2026 — Product homepage.
Further review notes & alternatives
Costs and operating effort
Open WebUI itself is free. For a local backend, prompts travel from the browser through the WebUI host to your local model service and stay there; for a hosted backend, they also leave for that provider and may incur that provider's own inference charges. The WebUI's own persistent storage (chats, users, settings) is a second data-retention responsibility independent of where the model itself runs, so "the model is local" doesn't mean "nothing is stored on the server."
Since v0.6.6, the license requires visible "Open WebUI" branding unless you qualify for an exemption: deployments of 50 or fewer users in any 30-day period can remove branding freely; substantive contributors can request written permission; and an enterprise license allows full white-labeling for a fee. Code contributed through v0.6.5 remains BSD-3-Clause with no branding requirement, so forking from that point avoids the restriction entirely, at the cost of missing later features.
Alternatives and trade-offs
LM Studio wins for a single desktop user who doesn't need a shared, multi-user, self-hosted interface.
Jan wins similarly for single-user local chat, with its own open-source licensing.
Langdock wins when you want a fuller team AI workspace (agents, reusable Skills, workflows) rather than primarily a chat front-end.
Give configured local or hosted model endpoints a self-hosted browser interface.
Timothy Jaeryang Baek — began Sept 2023 as "Ollama WebUI", renamed Open WebUI in Jan 2024 (Ollama raised a trademark concern); now community-maintained · free-tier · self-host
self-hosted chat
role-based access
multi-user interface
model integration
rag support
ldap sso
usage analytics
shared channels
notes workspace
talk-think
automate
search-cite
View sourced imageOfficial Open WebUI interface screenshot, unmodified with branding preserved.
open-webui/open-webui
Open WebUI License, not MIT. Redistribution notice and all branding retained; no endorsement is implied.
Open WebUI License
Copyright (c) 2023- Open WebUI Inc. [Created by Timothy Jaeryang Baek]
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Neither the name of the copyright holder nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
4. Notwithstanding any other provision of this License, and as a material
condition of the rights granted herein, licensees are strictly prohibited
from altering, removing, obscuring, or replacing any "Open WebUI"
branding, including but not limited to the name, logo, or any visual,
textual, or symbolic identifiers that distinguish the software and its
interfaces, in any deployment or distribution, except in the following
circumstances: (i) deployments or distributions where the total number
of end users (defined as individual natural persons with direct access
to the application) does not exceed fifty (50) within any rolling
thirty (30) day period; (ii) the licensee has obtained specific prior
written permission from the copyright holder; or (iii) where the
licensee has obtained a duly executed enterprise license expressly
permitting such modification. For all other cases, any removal or
alteration of the "Open WebUI" branding shall constitute a material
breach of license.
Materials governed by prior licenses retain those original license
terms, as specified in LICENSE_HISTORY.
By contributing to this project, you agree to the project's Contributor
License Agreement (CONTRIBUTOR_LICENSE_AGREEMENT).
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. Converted on 15 October 2026 to a 1600px-wide WebP (quality 80) for faster page loads; the original image remains in the media library.