HumanizerEvalSuggest a humanizerSuggest a tool

Leaderboard / Compare / Grubby vs Phrasly

October 2026 review · same 50 texts

Grubby vs Phrasly

In the October 2026 HumanizerEval review, Grubby and Phrasly rewrote the same 50 English texts. Grubby cleared all 6 detectors on 0 of them; Phrasly on 0. The difference is mostly Originality.ai Turbo: on 14 samples Grubby passed it where Phrasly failed, and the reverse never happened. On Pangram and Originality.ai 5.0 the two were close to even.

Per-sample agreement

Per-sample agreement, October 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 4.1Grubby4 to 1144
GPTZero 2026-10-07-baseGrubby15 to 31913
Originality.ai TurboGrubby14 to 0360
Originality.ai 5.0 model-5.0Tied0 to 0050
ZeroGPT defaultGrubby8 to 1401
Winston AI 5.0Grubby5 to 0450
All 6 detectorsTied0 to 0050

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 44 of 50 here.

Quality and output discipline

Quality and output discipline, October 2026.
MeasureGrubbyPhrasly
76%50%
64%70%
100%98%
78%84%
1.191.10
10
00
AcademicStandard model, API default (Phrasly Ultra not enabled on the account)
Websitegrubby.ai phrasly.ai

On per-sample quality scores, Grubby scored higher on 18 samples, Phrasly on 15, and the two tied on 17. Phrasly’s output was shorter on 45 of 50 samples; mean length ratios were 1.19 and 1.10, or 19% longer than the input and 10% longer than the input respectively.

Where each one outperforms the other

Grubby

  • Pangram: passed 5 of 50 against 2, including 4 samples where Phrasly failed, against 1 the other way.
  • GPTZero: passed 34 of 50 against 22, including 15 samples where Phrasly failed, against 3 the other way.
  • Originality.ai Turbo: passed 50 of 50 against 36, including 14 samples where Phrasly failed, against 0 the other way.
  • Originality.ai 5.0: tied at 0 of 50 but on a higher mean human score, 9.8% to 9.4%. It passed 0 samples where Phrasly failed.
  • ZeroGPT: passed 48 of 50 against 41, including 8 samples where Phrasly failed, against 1 the other way.
  • Winston AI: passed 50 of 50 against 45, including 5 samples where Phrasly failed, against 0 the other way.
  • Factual consistency: 76% against 50%.
  • Syntax integrity: 100% against 98%.

Phrasly

  • Pangram: passed 1 sample where Grubby failed, even though it was detected more often overall, passing 2 to 5.
  • GPTZero: passed 3 samples where Grubby failed, even though it was detected more often overall, passing 22 to 34.
  • ZeroGPT: passed 1 sample where Grubby failed, even though it was detected more often overall, passing 41 to 48.
  • Naturalness: 70% against 64%.
  • Stance preservation: 84% against 78%.
  • Length: shorter output on 45 of 50 samples, which matters under a word limit.

Grubby vs Phrasly FAQ

Is Grubby better than Phrasly?

In the October 2026 HumanizerEval review, Grubby cleared all 6 detectors on 0 of the 50 shared samples and Phrasly on 0. On Pangram and Originality.ai 5.0 the two were within 6 points of each other. Quality ran 18 samples to 15, with 17 ties.

Which humanizer is best for Pangram, Grubby or Phrasly?

Grubby. In October 2026, Grubby passed Pangram on 5 of the 50 shared samples and Phrasly on 2. On 4 samples Grubby passed where Phrasly failed; the reverse happened on 1. Both failed on 44.

Which humanizer changes the text less?

Grubby. It scored higher on per-sample quality on 18 of 50 samples, against 15. Grubby’s outputs averaged 1.19× the input length and passed factual consistency on 76%; Phrasly’s averaged 1.10× and passed on 50%. On stance preservation the split was 78% to 84%. Phrasly produced the shorter output on 45 of 50 samples.

More comparisons