GitHub Repository

Summarize any YouTube video in 12 languages using open-source LLMs without API keys.

11 starsPython

TubeTrim – A local YouTube summarizer using Qwen in pure Python

Name: TubeTrim – A local YouTube summarizer using Qwen in pure Python
Availability: InStock
Author: guglielmoce

by guglielmoce·Mar 9, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidCozySolve My Problem

Local Qwen 2.5-1.5B summarization when Glasp and Eightify already exist.

Strengths

•Hardware auto-detection prioritizes CUDA, then MPS, then CPU gracefully.
•Extractive compression handles transcripts exceeding context window limits.
•Zero API keys or subscriptions required for any functionality.

Weaknesses

•Depends on YouTube transcript availability, no audio-to-text fallback.
•1.5B model may struggle with complex or technical video content.

Post Description

I wanted a way to summarize YouTube videos without paying for a SaaS or leaking my viewing history to someone. TubeTrim is a Python-based tool that runs LLMs locally to process transcripts. No API keys, no subscriptions, no tracking.

It uses the transformers library with a device-aware backend: it will prioritize CUDA, then MPS (for Mac users), and finally fallback to CPU. I've found that Qwen 2.5-1.5B provides a good balance between speed and summary quality for this specific task.

How it works:

- Extracts the transcript via yt-dlp. - Performs extractive compression if the text exceeds the context window. - Summarizes via local inference with streaming output.

I'd appreciate any feedback for optimization!