I built a free auto-parts cross-reference API with a confidence score on every match

Building anything with car parts — a catalog, a stock lookup, a "compatible
alternatives" widget — you hit the same wall: cross-references are messy, and
every "analogs" API I'd used just hands you a flat list and says trust me. No
source, no confidence.

So I built one that doesn't. It's free on RapidAPI:

👉 Auto Parts Cross-Reference API

A confidence score on every cross

Each analog comes back with cf — a 0–100 confidence score derived from how
many independent sources agree — plus vu/vd (sources confirming / disputing).
So you can filter instead of guessing:

GET /v1/crosses?number=W%2075%2F3&min_confidence=70

Dirty part numbers just work

w75/3, W 75-3, W753 — send it however your user typed it, normalized
server-side:

curl "https://auto-parts-cross-reference-api.p.rapidapi.com/v1/crosses?number=w75/3" \
  -H "X-RapidAPI-Key: YOUR_KEY" \
  -H "X-RapidAPI-Host: auto-parts-cross-reference-api.p.rapidapi.com"

{ "r": [ {
  "p": { "pid": 1, "bid": 1, "b": "MANN-FILTER", "n": "W 75/3", "d": "Oil Filter", "oem": false },
  "a": [ { "p": { "pid": 2, "bid": 2, "b": "NISSAN", "n": "77119-45897", "oem": true }, "cf": 80, "vu": 1, "vd": 0 } ]
} ] }

Short field names keep payloads small (b=brand, n=number, d=description).
pid/bid are stable ids you can cache. Also on offer: /v1/numbers
(autocomplete), /v1/products?prefix=true (parts by prefix), /v1/crosses/by-id.

Why free

It's the focused piece of a bigger auto-parts data stack I'm building in public.
Crosses are useful on their own; if you later need vehicle fitment or the full
catalog, that's a separate API — you don't need it to get value here.

Free tier: 300 req/day, no card. →
Try it

If it breaks on your weirdest part number, tell me in the comments — that's half
the reason I'm shipping in the open.