Products Overview
There are many ways to get OpenAlex data, from point-and-click to whole-database. They all serve the same knowledge graph — the difference is how much you take, how fresh it is, how much technical skill it takes, and what it costs. This page helps you pick.
| Product | Best for | Technical skill | Cost |
|---|---|---|---|
| Website | Exploring, one-off questions, exporting result lists | None | Free |
| AI agent connector | Asking Claude in plain language to do most of what the website and API do; it can also fix your author profile. ChatGPT version coming soon | None | Free to add; queries use your own daily budget |
| Other agents | Setting up any other AI agent (ChatGPT, Cursor, coding agents) to query OpenAlex through the API | None | Free tier, then usage pricing |
| API | Apps, scripts, and analyses that need live data | Some coding | $1/day free, then usage pricing |
| CLI | Bulk downloads from your terminal, with retries and resume built in | Command line | Same as the API |
| Snapshot | Your own copy of the entire database, updated quarterly (release dates) | Data engineering | Free |
| Sync | Keeping your copy fresh with daily updates | Data engineering | Member+ and Partner plans |
| Fulltext | Full-text PDFs and TEI XML, per-file or the whole archive | Varies by option | $0.01/file; full archive via the PDF sync add-on |
| Unpaywall | Legacy Unpaywall integrations and OA lookups by DOI | Some coding | Free |
Picking a product
Just exploring? Use the website — search, filter, and export without writing a line of code. If you use Claude, the free OpenAlex connector lets you skip the clicking: ask in plain language and Claude runs the searches for you. Any other AI assistant can use OpenAlex too.
Building something? The API is the workhorse: fast, well-documented, and free for most uses — every account gets $1 of usage per day, which covers most research and personal projects. When you outgrow it, usage pricing is pay-as-you-go. The CLI wraps the same API for bulk work in your terminal, handling parallelism, checkpointing, and resume for you.
Want the whole database? The snapshot is the entire dataset, free to download, refreshed quarterly on the second Wednesday of January, April, July and October (release dates). If quarterly isn’t fresh enough, daily sync keeps you up to date. Keeping up to date is the paid service, part of the Member+ and Partner plans. The main route is a complete daily snapshot. For small numbers of updates, like just your own institution’s works, the API’s sync filters do the same job with no snapshot infrastructure at all.
After the documents themselves? Fulltext covers the content archive — more than 70 million open-access PDFs and their machine-readable TEI XML — from single-file downloads up to syncing the complete archive to your own bucket.
However you get it, all the data is free and open under CC0. What costs money is service: a bigger daily API budget, support, and keeping your copy up to date. That story, and every plan, lives in Pricing.
Last updated
View as Markdown