HumanizerEval

Leaderboard / Compare / Undetectable.ai vs RewriteAI

September 2026 review · same 50 texts

Undetectable.ai vs RewriteAI

In the September 2026 HumanizerEval review, Undetectable.ai and RewriteAI rewrote the same 50 English texts. Undetectable.ai cleared all 5 detectors on 15 of them; RewriteAI on 13. The difference is mostly ZeroGPT: on 6 samples Undetectable.ai passed it where RewriteAI failed, and the reverse never happened. On Pangram, Originality.ai and Winston AI the two were close to even.

Per-sample agreement

Per-sample agreement, September 2026. Counts are out of the 50 shared samples, computed at the 50% human-score threshold. Margin counts the samples only one tool passed, higher count first.
DetectorPassed more oftenMarginBoth passedNeither passed
Pangram 4.0Undetectable.ai8 to 6828
GPTZero 2026-09-13-baseUndetectable.ai6 to 1421
Originality.ai turboUndetectable.ai5 to 2421
ZeroGPT defaultUndetectable.ai6 to 0440
Winston AI 5.0Undetectable.ai2 to 1470
All 5 detectorsUndetectable.ai8 to 6729

Every count is computed from scores.csv at build time, not transcribed. “Neither passed” counts the texts that defeated both tools; on Pangram that was 28 of 50 here.

Quality and output discipline

Quality and output discipline, September 2026.
MeasureUndetectable.aiRewriteAI
26%54%
28%40%
94%100%
78%94%
1.211.09
72
00
University / General Writing / More Human, v11srAPI default, passages of up to 280 words (300-word API limit)
Websiteundetectable.ai rewriteai.com

On per-sample quality scores, Undetectable.ai scored higher on 9 samples, RewriteAI on 29, and the two tied on 12. RewriteAI’s output was shorter on 35 of 50 samples; mean length ratios were 1.21 and 1.09, or 21% longer than the input and 9% longer than the input respectively.

Where each one outperforms the other

Undetectable.ai

  • Pangram: passed 16 of 50 against 14, including 8 samples where RewriteAI failed, against 6 the other way.
  • GPTZero: passed 48 of 50 against 43, including 6 samples where RewriteAI failed, against 1 the other way.
  • Originality.ai: passed 47 of 50 against 44, including 5 samples where RewriteAI failed, against 2 the other way.
  • ZeroGPT: passed 50 of 50 against 44, including 6 samples where RewriteAI failed, against 0 the other way.
  • Winston AI: passed 49 of 50 against 48, including 2 samples where RewriteAI failed, against 1 the other way.

RewriteAI

  • Pangram: passed 6 samples where Undetectable.ai failed, even though it was detected more often overall, passing 14 to 16.
  • GPTZero: passed 1 sample where Undetectable.ai failed, even though it was detected more often overall, passing 43 to 48.
  • Originality.ai: passed 2 samples where Undetectable.ai failed, even though it was detected more often overall, passing 44 to 47.
  • Winston AI: passed 1 sample where Undetectable.ai failed, even though it was detected more often overall, passing 48 to 49.
  • Factual consistency: 54% against 26%.
  • Naturalness: 40% against 28%.
  • Syntax integrity: 100% against 94%.
  • Stance preservation: 94% against 78%.
  • Length: shorter output on 35 of 50 samples, which matters under a word limit.

Undetectable.ai vs RewriteAI FAQ

Is Undetectable.ai better than RewriteAI?

In the September 2026 HumanizerEval review, Undetectable.ai cleared all 5 detectors on 15 of the 50 shared samples and RewriteAI on 13. On Pangram, Originality.ai and Winston AI the two were within 6 points of each other. Quality ran 9 samples to 29, with 12 ties.

Which humanizer is best for Pangram, Undetectable.ai or RewriteAI?

Undetectable.ai. In September 2026, Undetectable.ai passed Pangram on 16 of the 50 shared samples and RewriteAI on 14. On 8 samples Undetectable.ai passed where RewriteAI failed; the reverse happened on 6. Both failed on 28.

Which humanizer changes the text less?

RewriteAI. It scored higher on per-sample quality on 29 of 50 samples, against 9. Undetectable.ai’s outputs averaged 1.21× the input length and passed factual consistency on 26%; RewriteAI’s averaged 1.09× and passed on 54%. On stance preservation the split was 78% to 94%. RewriteAI produced the shorter output on 35 of 50 samples.

More comparisons