What does "Unexpected token < in JSON at position 0" mean?

You almost certainly received an HTML error page (often <!DOCTYPE or <html) instead of JSON — a 404, 500, or a login redirect. Check the network tab for the actual response body and HTTP status before parsing.

How do I fix "Unexpected end of JSON input"?

The payload was cut off before the parser saw a closing brace or bracket. Common causes: aborted fetch, streamed response read incomplete, or res.json() called twice on the same body. Check Content-Length against the bytes you actually received.

Is it safe to paste production data here?

Yes. The tool is fully client-side — your JSON never leaves the browser tab.

Does it support JSON5 or relaxed JSON?

Yes — single quotes, comments, and unquoted keys are all auto-converted to strict JSON during repair.

Can it tell me which line the error is on?

Yes. The editor converts the position offset to a line/column and highlights the failing token directly.

Why does the error position seem wrong?

Some runtimes count differently when the input contains multi-byte UTF-8 characters. The fixer normalizes to character offsets, which matches what JSON.parse reports in modern V8 (Node 18+, Chrome 110+).

Fix "Unexpected Token in JSON" Errors

Paste the JSON that is throwing the error. We highlight the exact character position, explain what JSON.parse expected, and one-click repair the most common causes.

Indent

Format

Input

Waiting for input

Output

Instant

Runs entirely in your browser — no upload, no wait.

100% private

Your JSON never leaves the tab. Nothing is logged.

No signup

Free, unlimited, no account required.

What "Unexpected token" actually means

When JSON.parse reads your string, it walks character by character expecting a specific grammar: an object opens with a brace, every key is a double-quoted string, every value is one of the seven JSON types, and so on. The moment it sees a character that does not fit, it throws — and the position number is the byte offset where the parser gave up.

The message is generic on purpose. The same error can mean a missing quote, a stray comment, a Python-style True, a NaN, a single quote, or an actual control character. The fixer tells you which.

Common causes ranked by frequency

1. Single quotes instead of double quotes — copied from JavaScript or Python. 2. Unquoted keys — `{name: "Ada"}` is JS, not JSON. 3. Trailing commas at the end of arrays or objects. 4. Comments (// or /* */) — JSON has no comments. 5. Smart quotes — pasted from Word or Slack. 6. NaN, Infinity, undefined — none of these are valid JSON values. 7. Truncated payload — the response was cut off mid-string. 8. BOM or hidden characters — invisible bytes at the start of the file.

How to read the position number

The number after "position" is a zero-based character offset. So "Unexpected token N in JSON at position 42" means the parser failed at character 43. The editor in this tool jumps the cursor straight to that offset and highlights the surrounding line so you can see the context immediately.

try {
  JSON.parse(input);
} catch (e) {
  // e.message: 'Unexpected token } in JSON at position 42'
  const pos = +e.message.match(/position (\d+)/)?.[1];
  console.log('Failed near:', input.slice(Math.max(0, pos - 20), pos + 20));
}

Auto-repair vs manual fix

For most everyday issues — single quotes, trailing commas, comments, smart quotes — Smart Auto-Repair will hand you valid JSON in one click. For ambiguous cases (truncated payload, mismatched brackets) the tool surfaces the exact location and recommended fix without guessing.

Privacy and offline use

All parsing and repair happens locally in your browser. No upload, no server round-trip, no telemetry on the JSON itself. Works offline once the page is loaded.