spawnofsociety2/openwebui-openrouter-video — explained in plain English
Analysis updated 2026-05-18
Ask an OpenWebUI chat assistant to generate a short AI video from a text description without picking a model manually.
Generate a 4K cinematic video clip using a model like Veo 3.1 by describing the shot in natural language.
Use a reference image to guide the visual style of an AI-generated video within a chat conversation.
| spawnofsociety2/openwebui-openrouter-video | 1038lab/agnes-ai | adeliox/klein-head-swap | |
|---|---|---|---|
| Stars | 4 | 4 | 4 |
| Language | Python | Python | Python |
| Setup difficulty | easy | easy | moderate |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | developer | vibe coder | designer |
Figures from each repo's GitHub metadata at analysis time.
Requires an existing OpenWebUI instance and an OpenRouter API key, downloaded videos are never auto-deleted.
This project is a tool that adds AI video generation to OpenWebUI, a self-hosted chat interface for talking to language models. Once installed, a user's chat assistant can generate short AI videos just by being asked in plain language, without clicking through menus or picking a model manually. The assistant itself decides which video model to use, submits the generation job to OpenRouter, checks on it until it finishes, and then shows the finished video directly inside the chat. The tool pulls its list of available video models live from OpenRouter at the moment it is used, so it automatically supports new models like Sora, Veo, Kling, Seedance, Hailuo, Wan, and Grok as OpenRouter adds them, along with each model's supported resolutions, durations, aspect ratios, and pricing. Once a video finishes generating, the tool downloads the file to the OpenWebUI server itself rather than linking to a temporary URL, which avoids broken links after the original link expires, and embeds it in the chat with a working video player and download link. More advanced options include turning audio on or off where a given model supports it, using a fixed seed for repeatable results, and passing a reference image to guide the video's style. The README is notably careful about how audio behaves differently across models, since some models silently ignore the audio setting and produce sound anyway. Installing the tool means pasting a single Python file into OpenWebUI's tool editor and adding an OpenRouter API key in the tool's settings, a few optional settings control how long the tool waits for a video to finish before giving up. The project depends on the aiohttp Python library, which OpenWebUI installs automatically, and requires an existing OpenWebUI instance to run inside. One notable limitation mentioned in the README is that downloaded videos are never deleted automatically, so a long-running installation that generates many videos will need someone to periodically clear out old files. The project is released under the MIT license.
This OpenWebUI tool lets a chat assistant autonomously pick a video AI model on OpenRouter, generate a video from a plain-language request, and embed the finished clip directly in the chat.
Mainly Python. The stack also includes Python, aiohttp, OpenWebUI.
MIT license: use, modify, and distribute freely, including commercially, as long as the copyright notice is kept.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.