Nanosaur2-670M
well9472 · Kit
Makes illustrations from text prompts with a small 670M-parameter diffusion model, run through ComfyUI. Built as a low-budget research model for people who wan…
Edits video automatically with Claude as the director. It transcribes a clip, plans the edit, adds subtitles, zooms and sound effects, and exports an MP4. For creators who want a local dashboard.
AIEV is an open-source project on GitHub in the Productivity category, listed by Mindket. It is free. This is version 1, last updated on . Install it from GitHub as its README says. Get Mindket’s setup notes on your Mindket account, as a zip.
brew install ffmpeg
AIEV is an automatic video editor from noti.vn in which Claude directs the edit. You drop in a clip and describe what you want. The system transcribes it, plans the edit, builds animated text scenes, subtitles, zooms and sound effects, and exports an MP4. Besides editing, it can make a narrated video from an article, split a long video into shorts, and translate a video into other languages.
Creators and small teams who make a lot of talking-head or explainer video and want the AI to do the first edit while they supervise in a browser.
| Industry | Who uses it | How they use it | Example |
|---|---|---|---|
| Content creation | Solo creators | Turn a raw clip into an edited video with subtitles, zooms and sound effects | A creator drops in a talking-head clip and gets karaoke subtitles and beat-synced zooms |
| Publishing and media | Editors | Turn written articles into narrated videos | An editor pastes a web article and gets a spoken script and a narrated video project |
| Marketing and branding | Brand managers | Apply a brand kit with colors, fonts and a logo to every video | A brand kit puts the logo top-left on each output automatically |
| Localization | Translators | Translate the dialogue and add translated subtitles, a dubbed voice or both | A team converts a recorded talk into another language with dubbed audio |
Anyone who wants a hosted online service can skip it, because it runs on your own computer.
| Use case | Who | What they get back |
|---|---|---|
| AI video editing | Creators | An MP4 with scenes, subtitles, zooms and sound effects, drafted first and finalized later |
| Text to video | Publishers | A narrated video made from a URL or pasted text |
| Auto cut | Podcasters and speakers | Several shorts from one long video, split by time, highlights or your prompt |
| Translate video | Localization teams | Translated subtitles, a new voice track or both |
Several local tools need to be in place before the dashboard can edit anything.
On macOS you can install FFmpeg with this command.
brew install ffmpeg
The project includes an environment check that tests for each of these tools and installs what it is able to.
Open the dashboard, drop in a clip, describe the edit and click the button that starts editing with AI. A Guide button sits in the top bar of each page and opens a help panel for it.
| Tool | What it does |
|---|---|
| AI video editing | Analyzes the source, builds the scenes, assembles the timeline and exports an MP4 |
| Video editor | Previews the edit in the browser and lets you reorder, trim and split scenes on a timeline |
| Text to video | Writes a script from a URL or text, then narrates it and builds a video project |
| Auto cut | Splits one long video into many shorts |
| Translate video | Translates dialogue, then adds subtitles, a dubbed voice or both |
| Style Design | Keeps brand kits with colors, fonts, logo and tone and applies them to every output |
You drop in a long talking-head video and ask for short clips cut around the best moments. The system picks highlights, makes each one its own project, and all of them share one brief. You can then adjust any clip in the video editor, or ask the AI for changes in the chat panel.
AIEV is a video editor where an AI does the editing. You give it a clip and a short description, and it listens to the speech, plans the edit and adds titles, subtitles, zooms and sound effects. A first small thing to try is dropping in one short clip and starting an edit.
Underneath, Claude drives HyperFrames, which builds motion-graphics scenes from HTML and GSAP, and Remotion assembles them into a timeline. A Next.js dashboard on port 6868 talks to an Express backend on port 6869 over REST and server-sent events, with SQLite and a render queue. Every project keeps its state in one meta.json file, shared by the editor, the AI and the queue. The trade-off is a long list of local requirements.
# AIEV Edits video automatically with Claude as the director. It transcribes a clip, plans the edit, adds subtitles, zooms and sound effects, and exports an MP4. For creators who want a local dashboard. Made by notivn on GitHub. These notes are Mindket's, written from its README; the project's own files and license are in its repository. ## About What it is: AIEV is an automatic video editor from noti.vn in which Claude directs the edit. You drop in a clip and describe what you want. The system transcribes it, plans the edit, builds animated text scenes, subtitles, zooms and sound effects, and exports an MP4. Besides editing, it can make a narrated video from an article, split a long video into shorts, and translate a video into other languages. Who it suits: Creators and small teams who make a lot of talking-head or explainer video and want the AI to do the first edit while they supervise in a browser. ```table Industry | Who uses it | How they use it | Example Content creation | Solo creators | Turn a raw clip into an edited video with subtitles, zooms and sound effects | A creator drops in a talking-head clip and gets karaoke subtitles and beat-synced zooms Publishing and media | Editors | Turn written articles into narrated videos | An editor pastes a web article and gets a spoken script and a narrated video project Marketing and branding | Brand managers | Apply a brand kit with colors, fonts and a logo to every video | A brand kit puts the logo top-left on each output automatically Localization | Translators | Translate the dialogue and add translated subtitles, a dubbed voice or both | A team converts a recorded talk into another language with dubbed audio ``` Anyone who wants a hosted online service can skip it, because it runs on your own computer. Where to use it: - A web dashboard in your browser, served from your own machine - Premiere Pro and DaVinci Resolve, through XML export - macOS and NVIDIA machines, where hardware acceleration is detected automatically - Claude, which acts as the director through the Claude Agent SDK ```table Use case | Who | What they get back AI video editing | Creators | An MP4 with scenes, subtitles, zooms and sound effects, drafted first and finalized later Text to video | Publishers | A narrated video made from a URL or pasted text Auto cut | Podcasters and speakers | Several shorts from one long video, split by time, highlights or your prompt Translate video | Localization teams | Translated subtitles, a new voice track or both ``` How to set it up: Several local tools need to be in place before the dashboard can edit anything. - Node.js 24 LTS, or a recent 22 or 26 security release - FFmpeg on your PATH - Google Chrome, because the render tools use headless Chromium - Claude credentials, since Claude directs the edit - Python and faster-whisper for subtitles, and a Gemini key for Gemini voices and images On macOS you can install FFmpeg with this command. ```sh brew install ffmpeg ``` The project includes an environment check that tests for each of these tools and installs what it is able to. How to use it: Open the dashboard, drop in a clip, describe the edit and click the button that starts editing with AI. A Guide button sits in the top bar of each page and opens a help panel for it. ```table Tool | What it does AI video editing | Analyzes the source, builds the scenes, assembles the timeline and exports an MP4 Video editor | Previews the edit in the browser and lets you reorder, trim and split scenes on a timeline Text to video | Writes a script from a URL or text, then narrates it and builds a video project Auto cut | Splits one long video into many shorts Translate video | Translates dialogue, then adds subtitles, a dubbed voice or both Style Design | Keeps brand kits with colors, fonts, logo and tone and applies them to every output ``` Example: You drop in a long talking-head video and ask for short clips cut around the best moments. The system picks highlights, makes each one its own project, and all of them share one brief. You can then adjust any clip in the video editor, or ask the AI for changes in the chat panel. For newcomers and experts: AIEV is a video editor where an AI does the editing. You give it a clip and a short description, and it listens to the speech, plans the edit and adds titles, subtitles, zooms and sound effects. A first small thing to try is dropping in one short clip and starting an edit. Underneath, Claude drives HyperFrames, which builds motion-graphics scenes from HTML and GSAP, and Remotion assembles them into a timeline. A Next.js dashboard on port 6868 talks to an Express backend on port 6869 over REST and server-sent events, with SQLite and a render queue. Every project keeps its state in one meta.json file, shared by the editor, the AI and the queue. The trade-off is a long list of local requirements. ```flow You drop in a clip and describe the edit The AI transcribes it and plans the edit HyperFrames builds the motion-graphics scenes Remotion assembles the timeline The result exports as an MP4 ``` Good to know: - License: MIT. - Voices: 30 Gemini preset voices work online. - VieNeu-TTS runs offline and for free, and it is the one engine here that can clone your voice from a short recording. - Edit styles are off by default, and you can add your own to the 50 built in. - A brand logo that is missing is downloaded automatically, and logos are never drawn by AI. - Hardware acceleration uses NVENC on NVIDIA and VideoToolbox on macOS. ```chart Title: Built-in items by type Sound effects: 86 Brand logos: 125 Edit styles: 50 Gemini preset voices: 30 Color grading presets: 14 ``` ## On GitHub, 7 November 2026 - Stars: 124 - Forks: 62 - Open issues and pull requests: 5 - Language: TypeScript - License: MIT - Topics: ai-video-editor-sdk, claude, claude-agent-sdk, claude-ai, claude-code, claude-skills, edit-video, edit-videos, ffmpeg, hyperframes, motion-graphics, remotion ## Links - Code: https://github.com/notivn/AIEVv2 - Website: https://noti.vn/ - Releases: https://github.com/notivn/AIEVv2/releases - Issues: https://github.com/notivn/AIEVv2/issues - Star history: https://www.star-history.com/#notivn/AIEVv2&Date ## Get the code ``` git clone https://github.com/notivn/AIEVv2.git ``` Then follow the setup in its README. Nothing in this kit runs by itself.
README.md 6 KB
well9472 · Kit
Makes illustrations from text prompts with a small 670M-parameter diffusion model, run through ComfyUI. Built as a low-budget research model for people who wan…
Waqf-AI · Kit
Reads handwritten Arabic text from images and returns it as text. A fine-tuned PaddleOCR-VL model for people digitizing Arabic handwriting.
Wan-AI · Kit
Makes video clips from a text prompt or an image plus text, at up to 720P and 24 frames per second, in the Diffusers format for a single consumer GPU.
thedotmack · Kit
Alex Newman’s memory plugin for Claude Code and other coding agents. It records what the agent does in each session, turns that into short searchable notes and…
sirioberati · Skill
Gives you 74 animated explainer boards in HTML for YouTube videos, plus every prompt and three Claude skills that sync the boards to your voice and export an M…
heygen-com · Kit
HeyGen’s open-source framework that turns HTML into MP4 video. Its skills teach Claude Code, Codex, Cursor, Gemini CLI and other agents to plan, build, preview…