November 2026 · ZeroGPT Humanizer vs StealthGPT Super

ZeroGPT Humanizer Alternative: StealthGPT vs ZeroGPT

StealthGPT Super is the better alternative for the combined test results. Super passed 44 of 50 Pangram checks and ZeroGPT Humanizer passed 3. Super scored higher on all eight average writing-quality measures.

HumanizerEval · November 2026

StealthGPT Super vs ZeroGPT Humanizer on 50 matched inputs

The November 17 run compared 15 humanizers on the same 50 English inputs. Detector rows report passes on each detector. Quality rows report provisional automated scores.

StealthGPT Super

Super, Quality, rephrase enabled, API

ZeroGPT Humanizer

v11sr, University, General Writing, More Human, Business API

HumanizerEval November 2026 scores and strict passes for StealthGPT Super and ZeroGPT Humanizer
MeasureStealthGPT SuperZeroGPT Humanizer
Overall score70.34/100 (#2)30.76/100 (#7)
Weighted detector pass rate69.00%20.40%
Writing quality73.48/100 (#3)54.94/100 (#12)
Pangram 4.144/50 (88%)3/50 (6%)
GPTZero 5.0b30/50 (60%)15/50 (30%)
Originality.ai 5.019/50 (38%)0/50 (0%)
Winston AI 5.248/50 (96%)50/50 (100%)
ZeroGPT34/50 (68%)40/50 (80%)
Readability68.80/10046.00/100
Meaning preservation67.20/10051.60/100
Factual consistency78.00/10026.00/100

The overall score combines 70% weighted detector passes and 30% writing quality. Detector weights are 40% Pangram, 30% GPTZero, 20% Originality.ai, 5% Winston AI and 5% ZeroGPT. Winston uses a document Human Score of at least 80. The other detectors use qualifying human labels. HumanizerEval is operated by the StealthGPT team.

What the comparison shows

The biggest writing-quality gap was factual consistency. Super scored 78/100 compared with 26/100 for ZeroGPT Humanizer. In one matched example, ZeroGPT introduced a different name for the owner of a fictional bookstore. Super preserved the original name.

Read our ZeroGPT Humanizer review and matched examples

Frequently asked questions

Is ZeroGPT Humanizer the same product as the ZeroGPT detector?

ZeroGPT Humanizer rewrites text and the ZeroGPT detector classifies it. HumanizerEval tested the Business API rewrite model v11sr. Its rewrites passed ZeroGPT’s own checker 40 times and Pangram 3 times. GPTZero is a separate detector and passed 15 of the 50 ZeroGPT rewrites.

How did ZeroGPT Humanizer and StealthGPT Super compare on Pangram?

In November 2026, ZeroGPT Humanizer passed 3/50 (6%) on Pangram 4.1. StealthGPT Super passed 44/50 (88%) on the same 50 inputs. Only Pangram’s Human label counted as a pass.

What does the overall score measure?

HumanizerEval calculates the overall score using 70% weighted detector passes and 30% provisional automated writing quality. The quality rubric includes readability, factual consistency and six other measures. The scores apply to the configurations shown.