November 2026 · 750 outputs · 15 services
ZeroGPT detector benchmark
Model: Vendor-managed; no public version identifier. Weight in the humanization composite: 5%.
Unqualified Human Written or Most Likely Human written feedback passes. Mixed, uncertain and human labels qualified by possible AI fail. Human score is 100 minus fakePercentage, a text-share estimate. It is not GPTZero.
Small complementary signal; independent studies raise threshold and false-positive concerns. Account access does not establish calibration. Reference. None of these weights is a measured current detector accuracy.
| Humanizer | Passes/50 | Pass rate | Human-score average |
|---|---|---|---|
| Phrasly Ultra | 50/50 | 100.00% | 99.10 |
| WriteHuman | 46/50 | 92.00% | 93.78 |
| ZeroGPT Humanizer | 40/50 | 80.00% | 90.75 |
| Undetectable.ai | 39/50 | 78.00% | 90.44 |
| StealthGPT Super | 34/50 | 68.00% | 83.24 |
| Grubby | 22/50 | 44.00% | 70.99 |
| Rephrasy | 20/50 | 40.00% | 66.05 |
| BypassGPT | 20/50 | 40.00% | 66.88 |
| Humbot | 19/50 | 38.00% | 70.71 |
| Clever Humanizer | 18/50 | 36.00% | 51.37 |
| Solvely | 9/50 | 18.00% | 38.09 |
| RewriteAI | 7/50 | 14.00% | 34.44 |
| Grammarly Pro | 1/50 | 2.00% | 10.03 |
| QuillBot | 0/50 | 0.00% | 2.95 |
| HIX Bypass | 0/50 | 0.00% | 12.14 |
Failed strict bypass
100 minus the pass rate on these humanized outputs; this is not a false-positive rate on human writing.
| Humanizer | Failed/50 | Failed rate |
|---|---|---|
| Phrasly Ultra | 0/50 | 0.00% |
| StealthGPT Super | 16/50 | 32.00% |
| WriteHuman | 4/50 | 8.00% |
| Rephrasy | 30/50 | 60.00% |
| Undetectable.ai | 11/50 | 22.00% |
| Clever Humanizer | 32/50 | 64.00% |
| ZeroGPT Humanizer | 10/50 | 20.00% |
| BypassGPT | 30/50 | 60.00% |
| Grubby | 28/50 | 56.00% |
| Grammarly Pro | 49/50 | 98.00% |
| Humbot | 31/50 | 62.00% |
| Solvely | 41/50 | 82.00% |
| RewriteAI | 43/50 | 86.00% |
| QuillBot | 50/50 | 100.00% |
| HIX Bypass | 50/50 | 100.00% |
This measures bypass of the tested detector, not its false-positive rate or accuracy on independently verified human writing. A high pass rate alone cannot establish that the detector is bad.
Per-output scores CSV · Frozen inputs and prompts CSV · Scoring manifest · Verification bundle · GitHub report · Earlier editions