Why extract numbers for fact-checking?

Numeric claims are where LLMs hallucinate most confidently — a plausible but wrong percentage or dollar figure slips through easily. Pulling every number into a table with its surrounding claim makes it fast to spot and verify each one instead of re-reading the whole text.

What types of numbers are detected?

Percentages, currency amounts, plain counts and quantities, years and dates, multipliers like "3x", and numbers with units. Each is labelled by type so you can prioritize the high-risk statistics first.

Does it verify the numbers automatically?

No. It extracts and annotates them; verification against your trusted sources is up to you. This keeps it fully local and private — nothing is sent anywhere — while removing the tedious step of hunting for every figure manually.

Can it handle large pasted reports?

Yes. It runs in your browser and handles long text quickly. Each numeric match is tied back to the sentence it came from, so even in a long report you can jump straight to the claim that needs checking.

Number & Statistic Extractor

Find every number an LLM asserted, in one table

Language models state statistics with total confidence whether or not they are true. “Revenue grew 47% to $3.2M, beating the 12% industry average” reads authoritatively — and may be entirely invented. This tool scans LLM output, extracts every numeric claim with the sentence around it, and lays them out so you can fact-check fast.

How it works

The extractor matches several families of numbers: percentages, currency amounts, multipliers (“3x”), years, and plain counts, including those written with thousands separators or decimals. For each match it captures the surrounding sentence as the claim, labels the number by type, and puts it in a table. Everything runs locally in your browser, so you can paste confidential drafts safely.

Tips for fact-checking output

Check the high-impact figures first — percentages and currency amounts drive decisions, so verify those before counts.
Read the claim column, not just the number. A correct figure attached to the wrong subject is still a factual error.
Watch for suspiciously round or precise numbers. “Exactly 1,000,000 users” and “a 73.6% improvement” are both classic hallucination tells.
Pair with the date extractor to verify the temporal claims in the same text.