Thordata sells the usual proxy tiers, a couple of scraping APIs, and something most providers do not offer at all: ready made video datasets for AI work.
That last item says something about where this market is going.
The Proxy Layer
Residential addresses starting around sixty five cents a gigabyte, with ISP, datacenter and 4G and 5G mobile alongside.
Granular geographic targeting runs across the range, with sticky sessions, unlimited options, a developer oriented integration path and a free trial.
That is a complete and competitively priced catalogue on its own. The interesting part is what sits above it.
Two APIs Worth Separating
The SERP API returns structured search results from Google, Bing and others, in real time.
Search results deserve their own product because they are the hardest common target there is. They are personalised, localised, rate limited and reshuffled constantly. Building your own means maintaining parsers against a layout that changes without warning, which is a permanent part time job.
The Web Scraper API handles general collection, with rendering and unblocking so you get pages rather than challenges.
Our scraping guide covers what that maintenance actually involves, which is the honest way to judge whether buying it solved is worth the price for your work.
Video Datasets Are the Unusual Part
Selling finished datasets rather than the means to collect them is a meaningful step, and video specifically is harder than text.
Video is enormous. Collecting it at scale means bandwidth costs that make per gigabyte proxy pricing genuinely painful, plus storage, plus processing. For a lot of teams, buying a curated dataset is simply cheaper than building the pipeline that produces one.
It also reflects the obvious truth of this industry: nobody wants proxies. They want data, and proxies are a means to it. Providers selling the end result are skipping a step their customers never enjoyed anyway.
Escalate Through the Layers
The sensible way to use a provider structured like this is to buy the least you can get away with.
Raw proxies if your target is easy and you already have working extraction code. An API if the target is common and difficult and your parsers keep breaking. A dataset if the data already exists and collecting it yourself adds nothing.
Most people default to the bottom layer because it looks cheapest per unit, then spend months on maintenance that costs far more than the API would have.
Who It Fits
Thordata suits teams doing search results collection at scale, AI work needing video or large media datasets, anyone who wants to move between proxies and managed collection without changing vendor, and cost conscious buyers at the proxy layer.
Who Might Look Elsewhere
If you only need raw addresses, a pure proxy provider will be marginally cheaper and simpler.
Enterprise buyers wanting formal SLAs and named account management should look at the larger platforms.
Final Verdict
Thordata is a well rounded platform that sells the whole ladder from addresses up to finished data.
Four proxy tiers at competitive rates, a SERP API for the hardest common target, a general scraper API, and video datasets for teams who would rather buy than build.
Work out which layer your problem actually sits at before buying. Most people guess too low and pay for it in engineering time.
