Home/Answers/Big integers
JSON→TS / Big integers

How to Handle Big integers: 2026 Step-by-Step

To handle Big integers, confirm the input/format and output requirements, then follow the steps and verify the result.

Updated 2026-10-31About 3 minLocal test
Quick answer

To handle Big integers, confirm the input/format and output requirements, then follow the steps and verify the result. JSON numbers are IEEE754; large IDs may lose precision, so use strings.

Why it matters

Defining the output first avoids finishing before discovering the format/size/quality is wrong.

Above 2^53 is unsafe.

Step by step

  1. Start with real/complete JSON covering optional fields and empty arrays.
  2. Check array elements, null/optional and nested objects.
  3. Set the root name and copy into the project.
  4. Add Zod/Valibot runtime validation for production.
  5. Add contract tests/versioning for API changes.

Common mistakes

  • Treating a sample as a complete schema, missing optional/empty arrays.
  • Trusting generated any/null unions as-is.
  • Forgetting quotes for property names with spaces/Chinese.
  • Skipping runtime validation after generation.

Checklist

  • Sample covers optional/empty arrays/null.
  • Array object fields merged.
  • Property/root names valid.
  • Runtime validation added.
  • API contract/version considered.

Key facts

  • Above 2^53 is unsafe.
  • Backends should return strings.

FAQ

How do I handle Big integers?

To handle Big integers, confirm the input/format and output requirements, then follow the steps and verify the result. JSON numbers are IEEE754; large IDs may lose precision, so use strings.

How long does it take?

Usually 1-3 minutes; for intermittent issues, run 30 seconds and retest elsewhere.

Is data uploaded?

Runs locally and does not upload video/audio; extensions may still read the page.

Source: MDN Number