Skip to content

ProductFor teams shipping AI-drafted text

Fact-checking for any draft, before it goes out.

Lenz is a fact-checking API. It identifies factual statements in a draft, a memo, a model’s answer or a page of copy that it can check against independent public sources, and returns a result for each.

Try free Use it in Claude or ChatGPT

01Example

A paragraph a model wrote. One statement is wrong.

Example: a paragraph written by a model“GPT-4 was released in March 2023 with a context window of 128,000 tokens. OpenAI was founded in 2015 as a non-profit research laboratory.”

ClaimVerdictLenz ScoreSources
OpenAI released GPT-4 in March 2023. Truehigh confidence 10/10 go.nature.comreuters.comapnews.com+9 more
OpenAI was founded in 2015 as a nonprofit research laboratory. Truehigh confidence 10/10 storage.courtlistener.com990s.foundationcenter.orgtechcrunch.com+16 more
GPT-4 had a context window of 128,000 tokens at its release in March 2023.Suggested rewriteGPT-4 had a context window of 8,192 tokens at its release in March 2023. Falsehigh confidence 1/10 openai.comweb.archive.orgdevelopers.openai.com+10 more

Deep check · Checked 7 September 2026.

Open a row for the sources it used and the reasoning that weighed them.

Open a deep check to see which release the sources actually describe, and the reasoning that weighed them.

02Workflow

Use Lenz in your workflow.

  1. Hand over the text

    You hand over the text — a draft, a memo, an answer, a page of copy. Lenz reads it and picks out the factual statements it can check against public sources, so you are not listing them yourself. Each of those statements then gets a check.

  2. Choose a check per statement

    There are two kinds, and you choose per statement: a quick check is a first read, a verdict and a confidence level in about fifteen seconds, which is how you sort a long draft. A deep check researches independent public sources, has models argue both sides and a panel review them, and returns the verdict with its sources and reasoning in about a minute or two; if it changes the quick verdict, rely on the deep check.

  3. Use it where you work

    In the Workbench on a desktop, upload a Word document or paste a Google Doc link, and download the file with Lenz’s comments and suggested edits.

    Inside Claude, with one sign-in and no key; in Cursor and Claude Code, with one command and your key. From a script in Python or TypeScript. Use the REST API, n8n or Zapier to automate checks.

Developer docs

03Deep check

Deep check, in order.

  1. Framing

    Lenz makes the statement precise enough to test.

  2. Research

    It searches public sources for evidence on the statement.

  3. Debate

    Models from multiple vendors argue both sides using that evidence.

  4. Panel Review

    Three automated reviewers assess the arguments.

  5. Conclusion

    Lenz returns the verdict, the score, the sources and the reasoning trace.

That is what a deep check does, and it takes about a minute or two per statement. The example above is a paragraph Lenz wrote and checked; every row links to the published result.

Public sources only. Text only. Lenz does not judge whether the text was machine-written.

See the whole process

Warranty

Vendors disclaim accuracy. Lenz warrants the verdict.

A verdict can qualify when a full verification returns True or False with high panel confidence and enough independent sources behind it.

On a paid Pro or Scale plan, every full verification that qualifies carries a contractual warranty, at no extra charge.

How the warranty works

€10,000 warranty per qualifying fact check

04Pricing

Try free, no card.

Free

$0

100 quick checks or 10 deep checks a month, in any combination.

Try free
Plus

$7.99/ month

500 quick checks or 50 deep checks a month, in any combination.

Or $6.59 a month, billed $79 a year

Choose Plus
Pro · warranty included

$99/ month

5,000 quick checks or 500 deep checks a month, in any combination. The warranty on qualifying verdicts.

Or $83 a month, billed $990 a year

Choose Pro
Scale · warranty included

$399/ month

20,000 quick checks or 2,000 deep checks a month, for production volumes. The warranty on qualifying verdicts.

Or $333 a month, billed $3,990 a year

Choose Scale

Compare plans

Check a draft on the free tier.

Try free Use it in Claude or ChatGPT