October 2026 Benchmark Results
SmartHumanizer
Eighth overall in October at 60.4, with a relatively even profile across detector bypass (68.3), meaning (69.7) and readability (62.9).
Scorecard
By detector
How SmartHumanizer performs against each detector we test.
| Detector | Score | Pass rate |
|---|---|---|
| Pangram | 64.1 | 64% |
| GPTZero | 70.6 | 71% |
| Originality.ai | 69.8 | 70% |
| Copyleaks | 67.9 | 68% |
| ZeroGPT | 68.9 | 69% |
At a glance
SmartHumanizer sits 2.7 points behind Phrasly and 3.2 points ahead of GPTinf in the October overall ranking. Its three published components are relatively close, ranging from 62.9 to 69.7.
Detector results
SmartHumanizer places eighth on all five detector-specific rankings. It scores 64.1 against Pangram, 70.6 against GPTZero, 69.8 against Originality.ai, 67.9 against Copyleaks and 68.9 against ZeroGPT.
The trade-off
Meaning preservation at 69.7 is its strongest component, while readability at 62.9 is its weakest. None of the three reported components reaches the top half of the October field.
Bottom line
SmartHumanizer has a fairly even but lower-ranked October profile. It avoids the extreme quality trade-off seen in some detector specialists, but the published data does not show a category-leading strength.
October 2026 results
The benchmark case for SmartHumanizer
The current cycle places SmartHumanizer eighth overall with a relatively even spread across the three published components.
Overall: 60.4
Ranked #8 of 10 tools, 3.2 points ahead of ninth place.
Detector bypass: 68.3
Ranked #8 among the 10 October tools for detector bypass.
Meaning: 69.7
Ranked #8 among the 10 October tools for meaning preservation.
Readability: 62.9
Ranked #7 among the 10 October tools for readability.
Related research
Explore Humanizer Arena research
See separate experiments and the site methodology for additional context.