Make short videos without touching an editor

Name: Make short videos without touching an editor
Author: ATH-MaaS

The open-source project Pixelle-Video, by ATH-MaaS

“AI Fully Automated Short Video Engine” — the project's own words, on GitHub

What it does: Pixelle-Video turns a single topic into a finished short video, end to end, through a web page rather than a code editor. You type in a subject, and it writes the narration script, generates a matching AI image or video clip for each line, records the voice, adds background music, and assembles the whole thing into a video you can preview immediately. It can call hosted image and video services directly, or drive your own local ComfyUI setup if you already run one. You pick from ready-made visual templates for portrait, landscape, or square video, so the styling is chosen for you, not built from scratch.
Replaces: a short-form video production freelancer, or a stock-footage-and-editing subscription
For: For someone who wants a complete short video from one topic line, without touching a script, a storyboard, or an editing timeline.
Not for: Not for someone without either a local ComfyUI setup or a budget for image and video generation API calls — the free path needs a GPU capable of running ComfyUI, and the paid path needs image and video model credits.
Setup: 25 min · Claude Code, Kimi, Gemini CLI or Codex · uv (Python package manager), ffmpeg, an API key for an LLM to write the script, plus an image/video provider key or a local ComfyUI/RunningHub setup

Get it running

Paste this into the AI that runs on your computer. It does the install, checks it works, and tells you what to type first. If anything fails, paste the error back to it.

Paste into Claude Code

Install Pixelle-Video (https://github.com/ATH-MaaS/Pixelle-Video) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Claude Code, running on my machine.

1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password.
2. Clone https://github.com/ATH-MaaS/Pixelle-Video into ~/tools/pixelle-video and follow the README's install exactly.
3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says.
   Tool-specific notes: Install uv and ffmpeg first, and verify with uv --version and ffmpeg -version before cloning anything.
Launch it with uv run streamlit run web/app.py; it opens automatically at http://localhost:8501.
On first use, expand System Configuration and fill in an LLM key for scripts, then either a local ComfyUI URL (default http://127.0.0.1:8188, click Test Connection) or a RunningHub/API provider key for images and video — leave whichever you are not using blank.
If a rendered video looks static with no motion, check which Video Template group you picked (static_/image_/video_) — static templates never call an image or video model at all.
4. Run the smallest test the README gives, and show me the output.
5. When it works, tell me what to type first, in one line, for this job: Generate one AI-written video.
If anything fails, show me the exact error and fix it before going on.

Paste into Kimi

Install Pixelle-Video (https://github.com/ATH-MaaS/Pixelle-Video) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Kimi, running on my machine.

1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password.
2. Clone https://github.com/ATH-MaaS/Pixelle-Video into ~/tools/pixelle-video and follow the README's install exactly.
3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says.
   Tool-specific notes: Install uv and ffmpeg first, and verify with uv --version and ffmpeg -version before cloning anything.
Launch it with uv run streamlit run web/app.py; it opens automatically at http://localhost:8501.
On first use, expand System Configuration and fill in an LLM key for scripts, then either a local ComfyUI URL (default http://127.0.0.1:8188, click Test Connection) or a RunningHub/API provider key for images and video — leave whichever you are not using blank.
If a rendered video looks static with no motion, check which Video Template group you picked (static_/image_/video_) — static templates never call an image or video model at all.
4. Run the smallest test the README gives, and show me the output.
5. When it works, tell me what to type first, in one line, for this job: Generate one AI-written video.
If anything fails, show me the exact error and fix it before going on.

Paste into Gemini CLI

Install Pixelle-Video (https://github.com/ATH-MaaS/Pixelle-Video) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Gemini CLI, running on my machine.

1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password.
2. Clone https://github.com/ATH-MaaS/Pixelle-Video into ~/tools/pixelle-video and follow the README's install exactly.
3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says.
   Tool-specific notes: Install uv and ffmpeg first, and verify with uv --version and ffmpeg -version before cloning anything.
Launch it with uv run streamlit run web/app.py; it opens automatically at http://localhost:8501.
On first use, expand System Configuration and fill in an LLM key for scripts, then either a local ComfyUI URL (default http://127.0.0.1:8188, click Test Connection) or a RunningHub/API provider key for images and video — leave whichever you are not using blank.
If a rendered video looks static with no motion, check which Video Template group you picked (static_/image_/video_) — static templates never call an image or video model at all.
4. Run the smallest test the README gives, and show me the output.
5. When it works, tell me what to type first, in one line, for this job: Generate one AI-written video.
If anything fails, show me the exact error and fix it before going on.

Paste into Codex

Install Pixelle-Video (https://github.com/ATH-MaaS/Pixelle-Video) on this computer and get it working for me. I am not a developer; explain each step in one line as you go and never skip one. You are Codex, running on my machine.

1. Check what this machine already has (git, Python or Node as the project needs, Docker if the README says so). Tell me anything missing and install it, asking me before anything that needs my password.
2. Clone https://github.com/ATH-MaaS/Pixelle-Video into ~/tools/pixelle-video and follow the README's install exactly.
3. Configure it. If it needs an API key or a login, stop and ask me for it; never guess one and never store it anywhere except where the README says.
   Tool-specific notes: Install uv and ffmpeg first, and verify with uv --version and ffmpeg -version before cloning anything.
Launch it with uv run streamlit run web/app.py; it opens automatically at http://localhost:8501.
On first use, expand System Configuration and fill in an LLM key for scripts, then either a local ComfyUI URL (default http://127.0.0.1:8188, click Test Connection) or a RunningHub/API provider key for images and video — leave whichever you are not using blank.
If a rendered video looks static with no motion, check which Video Template group you picked (static_/image_/video_) — static templates never call an image or video model at all.
4. Run the smallest test the README gives, and show me the output.
5. When it works, tell me what to type first, in one line, for this job: Generate one AI-written video.
If anything fails, show me the exact error and fix it before going on.

What to point it at first

Generate one AI-written video

Type a simple topic into "AI Generated Content" mode and generate one full video end to end — the fastest way to confirm your LLM key, image/video provider, and TTS are all wired correctly.

Try a fixed script instead

Paste in a script you already wrote and use "Fixed Script Content" mode, so you can check the image and voice pipeline separately from the AI's own writing.

Preview before you commit

Use the Preview Style, Preview Voice, and Preview Template buttons before a full render — each is a cheap way to catch a wrong visual style or voice before spending a whole generation on it.

What it must never do unattended

Never publish a cut you have not watched to the end.
Check the licence of every model and every clip before it goes out under your name.
Never let it clone a voice from reference audio without the speaker's own permission — voice cloning is a built-in feature here, not a hypothetical.

Who made it

ATH-MaaS/Pixelle-Video on GitHub, under the Apache-2.0 licence. 27,914 stars, checked 9 September 2026. Last change 14 June 2026. We did not write it; we checked it, and wrote this page so you can use it.

Also in Content

Tool · Content · autoclip

One long video into a week of short clips

AutoClip watches a long video for you and pulls out the interesting parts.

20 min · your AI on your computer · docker, an API key for one AI provider (Qwen, an OpenAI-compatible API, Gemini, or SiliconFlow) — or a local Ollama/LM Studio install instead, with no key neededOpen

Tool · Content · ComfyUI

Make images and video without a designer

ComfyUI is a node-based studio for building AI image, video, and audio generation into a repeatable process, rather than typing one prompt at a time into a chat box.

30 min · your AI on your computer · a GPU (NVIDIA, AMD, Intel, or Apple Silicon — CPU-only works but is slow), python 3.12 or 3.13, your own model checkpoint files to load inOpen

Tool · Content · MoneyPrinterTurbo

Type a topic, get a finished short video

MoneyPrinterTurbo turns a topic into a finished short video without you touching an editing timeline.

20 min · your AI on your computer · python 3.11 or later, an API key for a script-writing AI model (or a local Ollama setup instead), ffmpeg (usually auto-downloaded, see brief)Open

What's in the library is how we work.

If you're launching or growing a brand across Asia and the West, a call is where we work out whether it's a fit.

Book a call Back to the library