November 2026 · Independent humanizer benchmarks. Explore the evidence

Compare AI humanizers

Pick two to four entries and inspect them side by side. All numbers come from the same shared source set and measurement profiles. An asterisk marks detector-dependent, provisional values.

Pick 2 to 4 entries 1 selected

November 2026
StealthGPTstealthgpt.ai
Closed source
Shared score components
Overall score36.44
Panel bypass score38.6%
Meaning preservation46.4%
Raw meaning cosine0.8644
Readability79.9%
Category consistency78.7%
Detector coverage231/231
Identical outputs0/33
Penalty points−12
Per-detector human-likelihood scores · measured outputs only
GPTZero12.6% 33/33
ZeroGPT15.6% 33/33
Copyleaks39.4% 33/33
Originality.ai9.3% 33/33
Pangram59.0% 33/33
Grammarly65.1% 33/33
QuillBot99.6% 33/33
Category scores · before entry-level penalties
Academic essay63.6
Application essay39.5
Blog post48.2
Business email36.7
Marketing copy47.4
Discussion board39.2
News article59.4
EvidenceAll 33 outputs