Leaderboard / Humanizers / Phrasly

October 2026 review · rank 7 of 11

Phrasly AI detector test results

In the October 2026 HumanizerEval review, Phrasly ranked 7 of 11 with a mean strict-bypass rate of 48.7%, behind StealthGPT at 80.3%. It cleared all 6 detectors on 0 of 50 texts. Review our full results for the best AI humanizer on the leaderboard page.

Phrasly results, October 2026

Website

phrasly.ai

Rank

7 of 11

Mean strict bypassThe average of the six detector pass rates: the share of texts each detector called human-written. The leaderboard is ranked by this number.

48.7%

Passed all 6 detectorsTexts that all six detectors called human-written.

0 of 50

Pangram 4.1 strict bypassHow many texts Pangram called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

2 of 50 (4%), mean 5.5%, median 0%

GPTZero 2026-10-07-base strict bypassHow many texts GPTZero called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

22 of 50 (44%), mean 45.2%, median 44.74%

Originality.ai Turbo strict bypassHow many texts Originality.ai Turbo called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

36 of 50 (72%), mean 66.8%, median 70.98%

Originality.ai 5.0 model-5.0 strict bypassHow many texts Originality.ai 5.0 called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

0 of 50 (0%), mean 9.4%, median 8%

ZeroGPT default strict bypassHow many texts ZeroGPT called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

41 of 50 (82%), mean 69.6%, median 76.4%

Winston AI 5.0 strict bypassHow many texts Winston AI called human-written (a human score above 50%), with the mean and median human score it gave across all texts.

45 of 50 (90%), mean 87.7%, median 99.53%

Factual consistency passShare of outputs where key claims, numbers, citations, and named entities remain accurate and correctly attributed.

50%

Naturalness passShare of outputs where wording is fluent and free of awkward or clunky phrasing.

70%

Syntax integrity passShare of outputs where sentences are grammatically complete and free of disruptive run-ons or punctuation errors.

98%

Stance preservation passShare of outputs where the source's position, tone, and intent are not reversed or distorted.

84%

Mean length ratioOutput word count divided by input word count, averaged across texts. 1.00 is the same length; 1.10 is 10% longer.

1.10

Samples over 1.4× inputOutputs more than 1.4× the input’s word count.

0

Samples under 0.6× inputOutputs shorter than 0.6× the input’s word count.

0

Settings usedThe mode each tool ran on: its strongest advertised setting on the cheapest paid tier.

Standard model, API default (Phrasly Ultra not enabled on the account)

Recompute any of this from scores.csv. Strict bypass means an overall human verdict on a human score above 50%.

Review history

Head to head

Because every tool saw the same texts, these pages report per-sample agreement, how often each tool bypassed detection on the same input, rather than a difference of averages.

• Phrasly vs StealthGPT

  Phrasly cleared all 6 detectors on 0 samples where StealthGPT did not; the reverse happened on 17. Both cleared 0.

• Phrasly vs Undetectable.ai

  Phrasly cleared all 6 detectors on 0 samples where Undetectable.ai did not; the reverse happened on 1. Both cleared 0.

• Phrasly vs WriteHuman

  Phrasly cleared all 6 detectors on 0 samples where WriteHuman did not; the reverse never happened. Both cleared 0.

• Phrasly vs Rephrasy

  Phrasly cleared all 6 detectors on 0 samples where Rephrasy did not; the reverse never happened. Both cleared 0.

• Phrasly vs Grubby

  Phrasly cleared all 6 detectors on 0 samples where Grubby did not; the reverse never happened. Both cleared 0.

• Phrasly vs RewriteAI

  Phrasly cleared all 6 detectors on 0 samples where RewriteAI did not; the reverse never happened. Both cleared 0.

• Phrasly vs HIX Bypass

  Phrasly cleared all 6 detectors on 0 samples where HIX Bypass did not; the reverse never happened. Both cleared 0.

• Phrasly vs Humbot

  Phrasly cleared all 6 detectors on 0 samples where Humbot did not; the reverse never happened. Both cleared 0.

• Phrasly vs BypassGPT

  Phrasly cleared all 6 detectors on 0 samples where BypassGPT did not; the reverse never happened. Both cleared 0.

• Phrasly vs Ryne

  Phrasly cleared all 6 detectors on 0 samples where Ryne did not; the reverse never happened. Both cleared 0.