- All features
- 10 reqs/minute
- Basic support
For: Testing and small experiments
Paste a link, get back clean media URLs and full metadata. No watermarks, no headless browsers, no scraping stack to maintain.
curl -s "https://api.saveapi.org/v1/download" \ -H "Authorization: Bearer sk_live_xxxx" \ -G --data-urlencode "url=https://instagram.com/reel/Cx1a2b3c"
{
"success": true,
"platform": "instagram",
"type": "video",
"medias": [
{ "type": "video", "ext": "mp4", "url": "…" }
],
"credits": { "spent": 1.5, "remaining": 998.5 }
}The response shape never changes between sources. Detect nothing, branch nowhere — write the parser once and forget it.
Everything is a GET. Authorize with a bearer token, pass the link, read the JSON.
/v1/downloadDetect the platform from the link and resolve it to direct media URLsfrom 1.5 credits/v1/{platform}Target one platform directly — same response shapefrom 1.5 credits/v1/youtube/infoYouTube: title, duration and every rendition with its exact size — no links2 credits/v1/youtube/createYouTube: build one rendition and return a signed download link10 credits/v1/instagram/profileInstagram: profile by username — name, bio, follower counts, avatar and recent posts3 credits/v1/detectCheck whether a link is supported before spending a creditfree/v1/platformsSupported platforms and the hosts each one ownsfree/v1/meKey details, plan limits and remaining creditsfree/v1/usageDaily request history and per-platform breakdownfreeBilled per request: 1.5 credits for a download, 5 for an Instagram story, 3 for Rutube. A link the source refuses — private or deleted — costs 0.1. Paid credits are valid for 30 days from purchase.
For: Testing and small experiments
For: Small apps and early-stage projects
For: Growing apps with stable traffic
For: High-volume production workloads
Message us on Telegram for a quick answer. Open a ticket from your dashboard when the problem needs a record — a failing request, a payment, a plan change.
For eleven of the thirteen platforms we never touch the files: the API returns direct links from the source's own CDN. The exceptions are YouTube — it signs its media URLs to the IP that requested them, so such a link is useless to you — and Twitch VODs, which Twitch publishes as HLS only; in both cases the file is streamed through us instead. Even then nothing is kept: the file streams through our server and is never written to disk.
You get a 4xx with a machine-readable error code, and the lookup costs 0.1 credits — a fifteenth of a download, because the work was done and came back empty. Bad links, unsupported sites, rate limits and our own failures cost nothing.
It depends on the platform and whether the source responds on the first try. We publish measured medians here only once there is enough traffic to report them honestly — we would rather show nothing than a number nobody measured.
Yes — that is the most common use. The direct URLs work with sendVideo and sendDocument, so you never touch the file yourself.
Per key, so each one carries its own limit. Create as many as you need from the dashboard and split traffic between your services or environments.
Key in under a minute. 1,000 free credits, no card, no expiry.