Loading page content
Pull structured fields out of unstructured text (emails, dates, amounts, phone numbers, identifiers) without writing a regex for each one.
Extract structured data from unstructured text using regex-based entity recognition. Finds emails, URLs, phone numbers, dates (multiple formats), IP addresses, currency amounts, and hex colors. Export results as a table, JSON, or CSV.
Paste text on the left to extract entities...
Extraction is where deterministic matching and model-based approaches genuinely split. Anything with a stable, checkable format (emails, IPs, UUIDs, ISO timestamps, currency amounts) is better served by a pattern than by a model. It is instant, free, and cannot hallucinate a value that was not in the input. Reach for a model only when the field is semantic, like the deadline mentioned in this email, where no pattern exists.
Dates are the standing exception in both directions. 03/04/2026 is ambiguous between two continents and no amount of pattern matching resolves it without knowing the source locale, which is exactly the kind of context a model can sometimes infer and a regex never can.