DEV Community

Ishwar Sirvi
Ishwar Sirvi

Posted on

Free public REST API for US sales tax rates by ZIP code

#api

I built and open-sourced a free public REST API for US sales tax lookups by ZIP code as part of ReceiptEdit — a free browser-based receipt generator.

Why

Every receipt-recreation tool I have seen makes users manually enter the tax rate. That is the single biggest reason rebuilt receipts fail an audit — a Manhattan restaurant tax rate is 8.875 percent, a Brooklyn one is different, a Queens one is different again. Manual entry gets it wrong at scale.

The endpoint 

GET https://receiptedit.com/api/sales-tax?zip=94103
Enter fullscreen mode Exit fullscreen mode

Returns the current 2026 combined state + county + city rate:

{
  "zip": "94103",
  "state": "CA",
  "county": "San Francisco",
  "city": "San Francisco",
  "combined_rate": 0.08625,
  "state_rate": 0.06,
  "county_rate": 0.0025,
  "district_rate": 0.02375
}
Enter fullscreen mode Exit fullscreen mode

No auth

No API key required. No account. No rate limit for reasonable use. Just a GET.

Iframe-embeddable widget

For sites that want to embed a live calculator instead of calling the JSON API:

<iframe src="https://receiptedit.com/api/embed/sales-tax/california" width="400" height="300"></iframe>
Enter fullscreen mode Exit fullscreen mode

Docs

Full endpoint documentation at receiptedit.com/developers. The underlying data updates when states publish new rate schedules — currently on the 2026 baseline.

What I use it for

The receipt generator uses it internally so every recreated receipt has a state-accurate tax line. If you are building anything similar — expense tools, freelancer bookkeeping, e-commerce checkout that needs a state rate lookup — feel free to use the API directly.

Related

MIT-friendly. Feedback welcome.Notes from open-sourcing https://github.com/receiptedit/us-sales-tax-2026 — the 2026 US sales tax dataset that powers https://receiptedit.com/sales-tax and the free public API at https://receiptedit.com/api/sales-tax.

  1. Real MIT license or dont bother. Awesome-list maintainers reject licenses with mandatory backlink clauses (correctly). Attribution appreciated, never required.

  2. Curated datasets get pickup that scraped data does not. Provenance matters.

  3. Shipping the same data three ways (raw files, REST API, iframe embed) captures three audiences. Different consumers pick the format that fits their stack.

  4. Cross-posting on 5 curated data lists produces more real traffic than 50 low-quality directories. Quality of the placement dominates.

  5. The dataset opened up B2B conversations the marketing site never did. Developers building tax-adjacent tools email you.

All the artifacts:

Top comments (0)