Videopython, a local-first Python library for video editing and AI workflows, is now available for download with both manual and LLM-driven editing modes.

Hacker News2h ago2 min read

Summaries like this, in your inbox every morning.

3 Key Points

What happened
A new Python library called videopython has been released for programmatic video editing, processing, and AI video workflows. It supports manual editing via JSON plans, automatic editing using a local AI model (Ollama), and integration with LLM agents through an MCP server. Installation includes core editing (pip install videopython), AI features (pip install "videopython[ai]"), and MCP server support (pip install "videopython[ai,mcp]").
Why it matters
The library runs AI features locally without requiring cloud API keys, meaning users can process videos and apply AI-driven editing on their own hardware. This addresses the need for privacy and cost control in video production workflows. The streaming architecture keeps memory bounded even for hour-long video sources, making it practical for longer content.
What to watch
The tool supports three AI integration modes—bringing your own LLM, using a bundled local vision model (gemma3:27b via Ollama), or running an MCP server for agent-driven editing. It requires Python >=3.11, <3.14, and a GPU is recommended for AI features. Full documentation is available at videopython.com.

No comments yet. Be the first to share your thoughts!

Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.

Free · takes 30 seconds · unsubscribe anytime

5 minutes a day. The AI essentials.

200+ sources · Email / LINE / Slack