Your last line of defense
against factual errors.

One credit equals one verified claim. When Vect scans your document, it identifies individual verifiable statements. Each statement you choose to verify costs one standard credit (or one deep research credit if you use Deep Research mode). Similar or duplicate claims are grouped automatically, so you are never charged twice for the same fact.

Vect uses semantic similarity to detect claims that express the same fact in different words. For example, “GDP grew 2.4%” and “The economy expanded by 2.4%” are recognized as the same claim and grouped together. You are only billed once per unique fact, regardless of how many times it appears in the document.

Standard mode performs fast verification against a broad index of trusted sources, typically completing each claim in about two seconds. Deep Research mode conducts a multi-source investigation across academic databases, government records, and specialized indexes — producing 8 to 13 times more evidence per claim. Deep Research is ideal for high-stakes documents such as legal briefs, clinical reports, and investment memos.

Grammarly makes sure you spelled “fraud” correctly. Vect makes sure you don’t accidentally commit it. Grammarly fixes your prose; we fix your facts.

You could, if you enjoy playing Russian Roulette with your reputation. General AI models are designed to sound confident, even when they hallucinate. Vect is the exact opposite. We hunt down primary sources, cross-reference the data, and give you the receipts. We verify.

Rest assured, we do exactly zero training on your uploads. Period. Your documents are encrypted, processed, and then we immediately forget they ever existed. You can keep your world domination plans to yourself.