ai detector.
Paste your draft and get one score with the exact sentences behind it, so you know how your writing reads before anyone else runs it through a checker.
See your result with a free account. No card required.
0 words
how it works.
01
Paste your draft
Drop in the full essay or the section you are unsure about. Longer samples give the model more to work with.
02
Read the marked lines
The report shows one score and highlights the sentences that influenced it most.
03
Put it in context
Read the sentence notes alongside your drafts, notes, and course policy before deciding what the result means.
Plenty of students run a check for the same reason: they wrote something themselves and still worry about how it reads. Stiff transitions, over-formal phrasing, and heavily templated structure are common patterns that can affect the estimate. Sentence-level notes show where the reading came from.
Lervan shows one score and the sentence-level signals behind it. Use the result as one input alongside your drafts, notes, and course policy.
what to check yourself
- The report shows exactly which lines were flagged and why, so you know what to revise before anyone else sees it.
- Private by default: your text is processed to return the result and is never saved to your account.
fair questions.
will this show me what my professor's checker will say?
No, and any tool that promises that is overpromising. Different checkers use different models. What this gives you is an early, sentence-level read on how your writing comes across, with time to revise before you submit.
what if my own writing gets a high score?
It happens, especially with formal academic prose. Read the sentence notes and compare the result with your drafts and notes. Your writing process is stronger evidence than a detector result.
can a score prove someone cheated?
No. Detection results are estimates, and no checker can prove who wrote something. Treat any score, from any tool, as one piece of context rather than a verdict.
how does the check work?
An AI model reads your text against the publicly documented signs of AI writing, including the "Signs of AI writing" guide maintained by Wikipedia editors and published research on words that became common in AI text after 2022. It returns one likelihood score and points to the sentences that shaped it, so you can see the reasoning instead of taking a number on trust.
is the ai detector free?
A Free account includes 5,000 AI-detection words each month. An eligible seven-day trial has its own 100,000-word allowance. Monthly and annual plans include 300,000 words each month, and the weekly plan 100,000 words each week.
Have your draft ready? Paste it in and check it now.
check yours now