November 2026 · 750 outputs · 15 services
Pangram detector benchmark
Model: 4.1. Weight in the humanization composite: 40%.
Native Human classification only; Mixed and AI fail. Human score is the fraction of text classified human, not document confidence.
Largest policy weight because independent research includes edited-AI detection evidence. This exact 40% is editorial, not a measured current accuracy. 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% | 100.00 |
| StealthGPT Super | 44/50 | 88.00% | 95.19 |
| WriteHuman | 33/50 | 66.00% | 76.92 |
| Rephrasy | 10/50 | 20.00% | 47.93 |
| Undetectable.ai | 6/50 | 12.00% | 36.35 |
| ZeroGPT Humanizer | 3/50 | 6.00% | 21.10 |
| Grubby | 1/50 | 2.00% | 8.68 |
| Solvely | 1/50 | 2.00% | 7.09 |
| Clever Humanizer | 0/50 | 0.00% | 6.48 |
| BypassGPT | 0/50 | 0.00% | 0.73 |
| Grammarly Pro | 0/50 | 0.00% | 0.00 |
| Humbot | 0/50 | 0.00% | 0.71 |
| RewriteAI | 0/50 | 0.00% | 6.30 |
| QuillBot | 0/50 | 0.00% | 0.00 |
| HIX Bypass | 0/50 | 0.00% | 0.00 |
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 | 6/50 | 12.00% |
| WriteHuman | 17/50 | 34.00% |
| Rephrasy | 40/50 | 80.00% |
| Undetectable.ai | 44/50 | 88.00% |
| Clever Humanizer | 50/50 | 100.00% |
| ZeroGPT Humanizer | 47/50 | 94.00% |
| BypassGPT | 50/50 | 100.00% |
| Grubby | 49/50 | 98.00% |
| Grammarly Pro | 50/50 | 100.00% |
| Humbot | 50/50 | 100.00% |
| Solvely | 49/50 | 98.00% |
| RewriteAI | 50/50 | 100.00% |
| 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