What Is the PatSnap API? Patent and Research Data for Developers
If you’re building patent search into a product, monitoring competitors, or feeding an R&D or AI workflow, you need patent and research data inside your code — not in a browser. Public sources make that hard: patent offices, literature databases, and chemical registries each live behind their own interface, with different formats and no shared schema. Stitching them together is a project before it’s a feature.
The PatSnap API solves that. It’s the set of REST endpoints on PatSnap Open Platform — the API platform built by PatSnap, the patent intelligence company — with programmatic access to 208M+ patents across 174 jurisdictions and 126M+ scientific papers, updated daily from WIPO, USPTO, EPO, and CNIPA. Everything comes back as structured data on one normalized schema.
What the PatSnap API Covers
The API spans patents through to life sciences and AI-powered analysis, organized into groups of endpoints so you call only what you need:
- Patent search — find patents by natural-language meaning, Boolean query, structured query, company name, or image.
- Patent data — pull full bibliographic records: abstracts, claims, descriptions, family, classifications, and citations.
- Legal status & value — legal status, litigation, transfers, licenses, fees, and valuation indicators.
- Scientific literature — search research papers and retrieve metadata, authors, journals, and patent citations.
- Drug, chemical & bio — drugs, clinical trials, targets and diseases; chemical structure search, ADMET prediction, and sequence alignment.
- AI endpoints — novelty search by technical feature, solution finding, technology autocomplete, and more.
That’s the core, not the whole list — the full endpoint catalog is on the API Reference. All of it shares one authentication scheme, one response format, and one set of standardized fields, so once you’ve made a single call, the rest work the same way.
Why Developers Use It
The data is already cleaned. Assignee names are standardized across offices, dates come in one consistent format, and legal status is refreshed daily. The normalization that usually eats your first sprint is done.
It keeps AI grounded. Because every result is a real value from an actual patent or paper record — publication numbers, assignees, dates — an LLM built on top cites grounded data instead of generating it, which keeps hallucinations down.
It’s easy to access. Standard REST: authenticate with a bearer token, send a query, read structured results back. No SDK, no per-office quirks.
It’s free to start. Every new account gets 10,000 credits with no credit card, enough to build and test a full integration before paying anything.
How to Get Started
Create an account on PatSnap Open Platform and open the Admin Dashboard to generate an API key — it’s issued once, in the sk- format, and you pass it as a bearer token in the Authorization header of every request. From there you can run a live call against any endpoint right in the interactive sandbox, before writing a line of integration code. The first request is usually a semantic patent search: you send a plain-language query and get back ranked patents as structured data.
The full endpoint catalog, parameters, response examples, and sandbox are on the API Reference page of PatSnap Open Platform.
Ready to Make Your First Call?
Access 208M+ patents across 174 jurisdictions and 126M+ scientific papers through one unified API. Start with 10,000 free credits — no credit card required.
→ Get Your API Key — sign up on PatSnap Open Platform
→ Explore the API Reference — full endpoint catalog and sandbox
Frequently Asked Questions
What is the PatSnap API?
It’s the set of REST APIs on PatSnap Open Platform that give programmatic access to patent, scientific-literature, and biomedical data — search, retrieval, legal status, and AI-powered analysis — across 174 jurisdictions, all on one normalized schema.
How do I get a PatSnap API key?
Create an account on PatSnap Open Platform, open the Admin Dashboard, and generate a key. It’s shown once in the sk- format, so copy it immediately and store it securely. Every new account comes with 10,000 free credits and no credit card is required.
How do I make my first request?
Pass your key as a bearer token in the Authorization header and send a query to an endpoint — for a first call, a semantic patent search takes a plain-language description and returns ranked patents. You can run it in the sandbox on PatSnap Open Platform before writing any code; the API Reference documents every parameter.
What data does the PatSnap API cover?
Patents (search and full bibliographic data, including legal status, citations, and valuation), scientific literature with patent citations, drug and biomedical data (drugs, clinical trials, targets, diseases), and chemical and bio data — plus AI-powered endpoints like novelty search and solution finding. Everything runs through one authentication scheme and response format.
Can AI agents connect through MCP?
Yes. Alongside the REST API, PatSnap Open Platform offers MCP (Model Context Protocol) servers, so assistants and agents like Claude can call patent search and retrieval as native tools without custom integration code. Browse them on the MCP Servers page of PatSnap Open Platform.
How much does the PatSnap API cost?
Every new account gets 10,000 free credits with no credit card required. Beyond that, it’s pay-as-you-go at $100 for 10,000 credits, so cost scales with usage rather than a fixed subscription.
