Odysee Scraper - Videos, Channels, Tags & Metadata
Pricing
from $2.00 / 1,000 items
Odysee Scraper - Videos, Channels, Tags & Metadata
Scrape Odysee (LBRY) videos and channels by search term or channel handle: title, description, channel, duration, tags, language, release date and thumbnail. No login, no proxy.
Pricing
from $2.00 / 1,000 items
Rating
0.0
(0)
Developer
Crawlyard Data
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
5 days ago
Last modified
Categories
Share
Odysee Scraper
Turns Odysee search terms and channel handles into clean rows of video metadata — title, description, channel, duration, tags, language, licence, release date and thumbnail.
No login, no API key, no proxy.
What one row looks like
{"title": "FREE TV FOR AS LONG AS IT LASTS!","url": "https://odysee.com/@SomeChannel:8/free-tv:342fd11c8a","claimId": "342fd11c8a...","channel": "@SomeChannel","channelTitle": "Some Channel","durationSec": 45,"releaseTime": "2026-09-18T01:46:49+00:00","tags": ["news", "politics"],"languages": ["en"],"license": "None","thumbnail": "https://thumbnails.odycdn.com/..."}
Input
| Field | Meaning |
|---|---|
searchTerms | Words or phrases to search for |
channels | @handle or a full odysee.com channel URL |
maxItemsPerQuery | Cap per term or channel, 1–1000 (default 50) |
orderBy | Newest, trending, most supported, or name |
streamType | Video, audio, or anything |
languages | ISO codes to restrict to, e.g. en, ko |
How it works
Odysee's own front end reads from an open JSON-RPC gateway. This Actor calls
the same claim_search and resolve methods directly, so a row costs a
fraction of what rendering the site would. The gateway caps a page at 50
claims, so larger requests are split into pages rather than quietly truncated.
Notes
- Only publicly published claims are read. Nothing private, no account needed.
releaseTimeis set by the uploader and can be faked. A minority of claims set it to the 32-bit maximum, which reads as the year 2038 and floats them to the top of "newest first". Those rows are flagged withreleaseTimeSuspicious: true.creationTimestampis the chain time and cannot be set by the uploader — sort on that when the date has to be real.- Paging walks a live index, so the same claim can appear on two pages. Rows
are deduplicated by
claimIdbefore they reach the dataset. - Videos without a thumbnail, licence or description return
nullfor those fields rather than a guess.