centersl/open-webui-docs
1
0
Fork 0
mirror of https://github.com/open-webui/docs.git synced 2026-03-27 13:28:37 +07:00
Code Issues Packages Projects Releases Wiki Activity

0.8.4

Browse Source
This commit is contained in:
DrMelone
2026-02-18 22:55:38 +01:00
parent 4e9a5011c0
commit f133aa922a
7 changed files with 99 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/features/ai-knowledge/prompts.md
View File

@@ -12,6 +12,7 @@ The Prompts interface provides several key features for managing your custom pro
* **Create**: Design new prompts with customizable names, tags, access levels, and content.
* **Share**: Share prompts with other users based on configured access permissions.
* **Access Control**: Set visibility and usage permissions for each prompt (refer to [Permissions](/features/access-security/rbac/permissions) for more details).
* **Enable/Disable**: Toggle prompts on or off without deleting them. Disabled prompts won't appear in slash command suggestions but remain in your workspace for future use.
* **Slash Commands**: Quickly access prompts using custom slash commands during chat sessions.
* **Versioning & History**: Track every change with a full version history, allowing you to compare and revert to previous versions.
* **Tags & Filtering**: Organize your prompts using tags and easily filter through your collection in the workspace.
@@ -293,6 +294,7 @@ Prompt management is controlled by the following permission settings:
* Document any specific requirements or expected inputs in the prompt description
* Test prompts with different variable combinations, including scenarios where optional fields are left blank
* Consider access levels carefully when sharing prompts with other users - public sharing means that it will appear automatically for all users when they hit `/` in a chat, so you want to avoid creating too many
* **Use the enable/disable toggle** to temporarily deactivate prompts you're not currently using instead of deleting them — this preserves the prompt's configuration and version history
* **Consider user workflows**: Think about which information users will always have versus what they might want to customize occasionally
### Migration Notes

77
docs/getting-started/quick-start/connect-a-provider/starting-with-anthropic.mdx Normal file
View File

@@ -0,0 +1,77 @@
---
sidebar_position: 3
title: "Anthropic (Claude)"
---
## Overview
Open WebUI supports **Anthropic's Claude models** natively through their OpenAI-compatible endpoint. Just plug in your API key and start chatting — no middleware, no pipes, no extra setup required.
Open WebUI includes a built-in compatibility layer that automatically detects Anthropic URLs, handles model discovery, and translates requests. Chat completions, streaming, and **tool calling** all work correctly out of the box.
:::warning Anthropic's Compatibility Disclaimer
Anthropic states that their OpenAI-compatible API is intended **for evaluation and testing**, not production workloads. In practice, everything — including tool calling — works as intended, but be aware of this official stance. For full Claude-native features (PDF processing, citations, extended thinking, prompt caching), Anthropic recommends their native `/v1/messages` API, which can be accessed via a [pipe function](/features/extensibility/plugin/functions/pipe) or a proxy like [LiteLLM](https://docs.litellm.ai/).
:::
---
## Step 1: Get Your API Key
1. Go to [console.anthropic.com](https://console.anthropic.com/).
2. Create an account or sign in.
3. Navigate to **API Keys** and generate a new key.
---
## Step 2: Add the Connection in Open WebUI
1. Open Open WebUI in your browser.
2. Go to ⚙️ **Admin Settings** → **Connections** → **OpenAI**.
3. Click ➕ **Add Connection**.
4. Enter the following:
| Setting | Value |
|---|---|
| **URL** | `https://api.anthropic.com/v1` |
| **API Key** | Your Anthropic API key |
5. Click **Save**.
:::tip
The URL field will suggest common provider endpoints as you type — just select the Anthropic URL from the dropdown.
:::
---
## Step 3: Start Chatting!
That's it! Open WebUI automatically detects the Anthropic endpoint and fetches available models. Select a Claude model from the model dropdown and you're good to go.
Streaming, multi-turn conversations, and tool calling all work seamlessly.
:::info Optional: Filter Models
If you want only specific models to appear, you can add model IDs to the **Model IDs (Filter)** allowlist in the connection settings. Common model IDs:
| Model | ID |
|---|---|
| Claude Sonnet 4.5 | `claude-sonnet-4-5-20250514` |
| Claude Opus 4 | `claude-opus-4-0-20250514` |
| Claude Sonnet 4 | `claude-sonnet-4-20250514` |
| Claude Haiku 3.5 | `claude-3-5-haiku-20241022` |
Check [Anthropic's model documentation](https://docs.anthropic.com/en/docs/about-claude/models) for the latest available model IDs.
:::
---
## Managing Your Connection
You can **enable or disable** your Anthropic connection at any time using the toggle switch next to the connection entry. This lets you temporarily deactivate a provider without deleting the connection or losing your configuration.
---
## Learn More
- **[OpenAI-Compatible Providers](/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible)** — Protocol design philosophy and setup for other providers.
- **[Pipe Functions](/features/extensibility/plugin/functions/pipe)** — Use Anthropic's native API with full feature support via a pipe function.

20
docs/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible.mdx
View File

@@ -40,7 +40,7 @@ When you add a connection, Open WebUI verifies it by calling the provider's `/mo
| Provider | `/models` works? | Action Needed |
|---|---|---|
| Anthropic | No — uses non-standard auth headers | Add model IDs manually to the whitelist |
| Anthropic | Yes — built-in compatibility layer | Auto-detection works |
| GitHub Models | No — uses non-standard path | Add model IDs manually to the whitelist |
| Perplexity | No — endpoint doesn't exist | Add model IDs manually to the whitelist |
| MiniMax | No — endpoint doesn't exist | Add model IDs manually to the whitelist |
@@ -61,7 +61,7 @@ When you add a connection, Open WebUI verifies it by calling the provider's `/mo
1. Open Open WebUI in your browser.
2. Go to ⚙️ **Admin Settings** → **Connections** → **OpenAI**.
3. Click ➕ **Add Connection**.
4. Fill in the **URL** and **API Key** for your provider (see tabs below).
4. Fill in the **URL** and **API Key** for your provider (see tabs below). The URL field will **suggest common provider endpoints** as you type.
5. If your provider doesn't support `/models` auto-detection, add your model IDs to the **Model IDs (Filter)** allowlist.
6. Click **Save**.
@@ -69,22 +69,26 @@ When you add a connection, Open WebUI verifies it by calling the provider's `/mo
If running Open WebUI in Docker and your model server is on the host machine, replace `localhost` with `host.docker.internal` in the URL.
:::
:::tip Enable/Disable Connections
Each connection has a **toggle switch** that lets you enable or disable it without deleting the connection. This is useful for temporarily deactivating a provider while preserving its configuration.
:::
### Cloud Providers
<Tabs>
<TabItem value="anthropic" label="Anthropic" default>
**Anthropic** (Claude) offers an OpenAI-compatible endpoint. Note that this is intended for testing and comparison — for production use with full Claude features (PDF processing, citations, extended thinking, prompt caching), Anthropic recommends their native API.
:::tip
See the dedicated **[Anthropic (Claude)](/getting-started/quick-start/connect-a-provider/starting-with-anthropic)** guide for a full step-by-step walkthrough.
:::
**Anthropic** (Claude) offers an OpenAI-compatible endpoint. Open WebUI includes a built-in compatibility layer that automatically detects Anthropic URLs and handles model discovery — just plug in your API key and models are auto-detected. Note that this is intended for testing and comparison — for production use with full Claude features (PDF processing, citations, extended thinking, prompt caching), Anthropic recommends their native API.
| Setting | Value |
|---|---|
| **URL** | `https://api.anthropic.com/v1` |
| **API Key** | Your Anthropic API key from [console.anthropic.com](https://console.anthropic.com/) |
| **Model IDs** | **Required** — add manually (e.g., `claude-sonnet-4-5-20250514`, `claude-opus-4-0-20250514`) |
:::caution
The `/models` endpoint **will fail** for Anthropic because their API uses non-standard authentication headers (`x-api-key` instead of `Bearer`). This is a cosmetic issue — **chat completions will work correctly**. Just add your model IDs to the allowlist manually.
:::
| **Model IDs** | Auto-detected — leave empty or filter to specific models |
</TabItem>
<TabItem value="gemini" label="Google Gemini">

4
docs/getting-started/quick-start/connect-a-provider/starting-with-openai.mdx
View File

@@ -8,7 +8,7 @@ title: "OpenAI"
Open WebUI makes it easy to connect to **OpenAI** and **Azure OpenAI**. This guide will walk you through adding your API key, setting the correct endpoint, and selecting models — so you can start chatting right away.
For other providers that offer an OpenAI-compatible API (Anthropic, Google Gemini, Mistral, Groq, DeepSeek, and many more), see the **[OpenAI-Compatible Providers](/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible)** guide.
For other providers that offer an OpenAI-compatible API (Google Gemini, Mistral, Groq, DeepSeek, and many more), see the **[OpenAI-Compatible Providers](/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible)** guide. For Anthropic's Claude models, see the dedicated **[Anthropic (Claude)](/getting-started/quick-start/connect-a-provider/starting-with-anthropic)** guide.
---
@@ -107,7 +107,7 @@ Simply choose GPT-4, o3-mini, or any compatible model offered by your provider.
That's it! Your OpenAI API connection is ready to use.
If you want to connect other providers (Anthropic, Google Gemini, Mistral, Groq, DeepSeek, etc.), see the **[OpenAI-Compatible Providers](/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible)** guide.
If you want to connect other providers, see the **[Anthropic (Claude)](/getting-started/quick-start/connect-a-provider/starting-with-anthropic)** guide or the **[OpenAI-Compatible Providers](/getting-started/quick-start/connect-a-provider/starting-with-openai-compatible)** guide for Google Gemini, Mistral, Groq, DeepSeek, and more.
If you run into issues or need additional support, visit our [help section](/troubleshooting).

4
docs/getting-started/quick-start/tab-docker/DockerUpdating.md
View File

@@ -5,7 +5,7 @@ To update your local Docker installation to the latest version, you can either u
### Option 1: Using Watchtower (Recommended Fork)
:::info Deprecation Notice
The original `containrrr/watchtower` is **no longer maintained** and may fail with newer Docker versions. We recommend using the `nickfedor/watchtower` fork.
The original `containrrr/watchtower` is **no longer maintained** and may fail with newer Docker versions. We recommend using the `nicholas-fedor/watchtower` fork.
:::
:::warning Multi-Worker Environments
@@ -17,7 +17,7 @@ If you run Open WebUI with `UVICORN_WORKERS > 1` (e.g., in a production environm
3. Stop and restart the container with your desired number of workers.
:::
With [Watchtower](https://github.com/nickfedor/watchtower), you can automate the update process:
With [Watchtower](https://github.com/nicholas-fedor/watchtower), you can automate the update process:
```bash
docker run --rm --volume /var/run/docker.sock:/var/run/docker.sock nickfedor/watchtower --run-once open-webui

6
docs/getting-started/updating.mdx
View File

@@ -239,9 +239,9 @@ Automated updates can save time but require careful consideration of the trade-o
The original `containrrr/watchtower` is **no longer maintained** and **does not work with Docker 29+**. The community has created maintained forks that resolve these issues.
:::
The original Watchtower project hasn't received updates in over two years and fails with Docker version 29.0.0 or newer due to API version incompatibility. Two maintained forks are now available: nickfedor/watchtower and Marrrrrrrrry/watchtower, both compatible with Docker 29+.
The original Watchtower project hasn't received updates in over two years and fails with Docker version 29.0.0 or newer due to API version incompatibility. Two maintained forks are now available: nicholas-fedor/watchtower and Marrrrrrrrry/watchtower, both compatible with Docker 29+.
**Recommended: nickfedor/watchtower fork**
**Recommended: nicholas-fedor/watchtower fork**
<Tabs>
<TabItem value="one-time" label="One-Time Update" default>
@@ -453,7 +453,7 @@ For complete Diun documentation, visit https://crazymax.dev/diun/
| **Best For** | Set-and-forget homelabs | Visual monitoring + control | Notification-only workflows |
:::tip Recommendation
- **For homelabs/personal use:** nickfedor/watchtower (automated)
- **For homelabs/personal use:** nicholas-fedor/watchtower (automated)
- **For managed environments:** WUD (visual + manual control)
- **For production/critical systems:** Diun (notifications only) + manual updates
:::

2
docs/intro.mdx
View File

@@ -151,7 +151,7 @@ Always use a separate data volume (e.g., `-v open-webui-dev:/app/backend/data`)
To update Open WebUI container easily, follow these steps:
#### Manual Update
Use [Watchtower](https://github.com/nickfedor/watchtower) to update your Docker container manually:
Use [Watchtower](https://github.com/nicholas-fedor/watchtower) to update your Docker container manually:
```bash
docker run --rm -v /var/run/docker.sock:/var/run/docker.sock nickfedor/watchtower --run-once open-webui
```