TL;DR: Free China-format test data generator: resident ID numbers (MOD 11-2 valid), mobile numbers (real prefixes), Unified Social Credit Codes (mod 31), and bank cards (Luhn + public test BINs) in batches. Format-valid yet random — mapping to no real person or account. Runs locally.
This tool runs entirely in your browser. Open DevTools → Network panel and search your input — it appears in no request.
Usage Guide
Load tests need 500 valid ID numbers, form integration needs phone numbers that pass validation, and sometimes you just want to exercise Luhn locally — test data is a dev routine. This tool batch-generates four China formats: everything passes its national-standard checksum (verify with this site’s validators), while every digit is random and maps to no real person, company, or account. Fully local.
1. What format-valid-but-random means
ID numbers pass MOD 11-2, phones use real prefixes, credit codes pass mod 31, cards pass Luhn — they fool format validation, which is exactly the job of test data. But sequences are random: the odds of colliding with a registered number are astronomically small, and nothing is paired with names (this tool deliberately never generates name-number pairs). Bank cards use public payment-gateway test BINs (411111 etc.) that fail in real payment environments by design.
2. Compliance boundaries (read this)
- Allowed: dev/test environment fill, form-validation integration, teaching demos, replacement data for anonymization drills.
- Forbidden: impersonating anyone on real services, defeating real-name verification, or feeding phishing/fraud pages — the Cybersecurity Law and PIPL attach clear liability.
- Gray-line test: if generated data would leave your test environment and touch a real system or a real person, stop.
3. Pair it with the validators
The recommended engineering loop: generate a batch → feed your form → point the form at the same rules this site’s validators use → deliberately corrupt one digit and confirm the validator catches it. The loop surfaces both failure classes — checks too loose (corrupted data passes) and too strict (valid data rejected) — and the latter, in production, locks out real users, which is usually the costlier bug.
Related Tools
Related Articles
China Generative AI Filing Checklist: Algorithm Registration vs. Large-Model Launch Filing
Offering generative AI services to the public in China means passing two filings: the algorithm registration under the Deep Synthesis Provisions and the large-model launch filing under the Interim Measures for Generative AI Services. This post maps the trigger conditions, the materials framework, the corpus and security-assessment pain points, and a self-check order. Practical guidance; always defer to the latest CAC templates.
MLPS 2.0 (China's Cybersecurity Multi-Level Protection Scheme) Self-Check: GB/T 22239, Layer by Layer
A practical self-check framework before China's MLPS level-protection evaluation (dengbao ceping): the filing workflow, high-frequency items across the five technical and five management layers, level-2 vs. level-3 differences, how MLPS relates to the cryptographic assessment (miping), and a remediation order ranked by points-per-effort.
TOTP Two-Factor Authentication Complete Guide: Protect Your Account Security
Deep understanding of how TOTP (Time-based One-Time Password) works, master the usage and best practices of two-factor authentication