The five seconds between paste and download
You paste a link. The page reads it, checks the shape locally, and sends just that URL to our server. Nothing else goes with it: no clipboard history, no page you came from, no account information, because there is no account.
The server pulls the post ID out of the link and asks Instagram's public endpoints for the media attached to it. These are the same endpoints your browser uses when it renders an embedded post on a news site. If one is unavailable, it tries the next; there are several, because Meta changes them often.
What comes back is a description of the post: which files exist, what resolution each one is, who posted it. That gets tidied into the list of buttons you see. When you click one, the file streams through our server to your device and is gone the moment the transfer finishes.
Why the file passes through us at all
It would be simpler to link you straight to Instagram's CDN. Three things stop that working properly.
The CDN sends no Content-Disposition header, so a direct link opens the
video in a new tab instead of saving it; on iOS especially, that is a dead end. It
sends no CORS headers either, so the browser will not let a script fetch it. And the
filename would be a forty-character hash rather than something you can find again.
So the file is streamed through, in chunks, adding the headers that make a download behave like a download. The common alternative (loading the whole video into memory as a blob before saving it) works fine for a ten-second clip and falls over on a phone with a long video. Streaming costs nothing on your side.
What is stored
The resolved description of a post is cached for fifteen minutes, so that a link being shared around does not trigger a fresh lookup every time. It holds the media URLs and the caption. It does not hold your IP address, and it expires on its own.
Media files are never written to disk. They pass through memory in 64 KB chunks and are not retained. There is no download history, because there is nothing to attach one to.
Rate limiting counts requests per IP address in a rolling window, which is the one place an IP is used. Those counters expire within the minute.
What it cannot do
It cannot reach private accounts. This is worth repeating because so many sites imply otherwise: private content is refused by Instagram before any request of ours gets near it. Following the account makes no difference, because the tool is not logged in as you and never will be.
It cannot recover expired stories, deleted posts, or archived content. It cannot produce resolution that was never uploaded. And it cannot remove a watermark that was burned into the video frames before the creator posted it; hold on to a clip you want to study later explains that distinction properly.
When it breaks
It will, occasionally. Instagram changes its public interfaces without warning and without any obligation not to. When that happens, lookups start failing until the resolver is updated. Any tool in this category that claims permanent reliability is describing something that does not exist.
If a link fails, the first thing to check is whether you can open it logged out. That single test separates "the account is private" from "the tool is broken", and it is right about nine times out of ten.
Questions people actually ask
Do you log what I download?
No. There is no per-user history stored anywhere, because there are no user accounts to attach one to.
Do you keep the video files?
No. They stream through memory and are not written to disk at any point.
Is my Instagram account involved?
Not at all. You are never asked to log in, and nothing here touches your account.