ModelAny
中文 Install extension

Video & page guide · 2026-09-28

Summarize YouTube and Bilibili videos with your AI accounts

Attach a transcript, then ask the AI you already use

On a YouTube or Bilibili watch page, open ModelAny and choose Attach transcript. Captions are requested from that video site in your current session—not scraped through a ModelAny proxy. Segments keep timestamps (about 30 seconds each) so answers can cite moments in the video.

The same popup can Attach text on ordinary articles (navigation and ads stripped) or attach a local PDF / TXT / Markdown / HTML file parsed in the browser. After attaching, send a ready-made prompt—Key points, Chapters, Quotes, action items, find flaws, translate—or write your own.

Summarize a video

  1. Open the video on YouTube or Bilibili and wait until the player is ready
  2. Open ModelAny → Attach transcript (Chinese or English captions when available)
  3. Pick a one-tap prompt such as Key points or Chapters, or type your own question
  4. Choose one or more AI sites and send; review answers with timestamp references

Articles and local files

Attach text works on typical article pages. Local PDFs and text files are parsed on your device; the file is not uploaded to ModelAny. Length limits still apply on each AI website—trim or ask for a section-by-section summary when the source is very long.

Accuracy and limits

Quality depends on the captions or article text the site provides. Auto-generated captions can mishear names and jargon. ModelAny does not invent a transcript when the video has none.

Get the extension

Install from the official store for your browser. Drafts, settings, history and memory stay on your device; prompts go only to the AI sites you select.

Chrome Web Store · Microsoft Edge Add-ons

Related pages

Frequently asked questions

Does this work without a YouTube premium plan?

Yes. It uses captions the watch page already exposes to your browser session. Availability depends on the video having captions, not on a ModelAny subscription.

Are captions sent to ModelAny servers?

No. Captions are read in-page from YouTube or Bilibili. They leave your machine only when you send a prompt to an AI site you select.

Can I summarize Bilibili videos?

Yes. Bilibili is supported the same way as YouTube: Attach transcript on the watch page, then ask your chosen models.

Summarize videos with your own AI accounts

Install ModelAny, attach a transcript on YouTube or Bilibili, and send. No API key, no ModelAny credits. Chrome Web Store · Microsoft Edge Add-ons

Install extension
Install extension