October 2026 Benchmark Results
GPTinf
Ninth overall in October at 57.2, with 65.6 for detector bypass, 67.4 for meaning preservation and 58.1 for readability.
Scorecard
By detector
How GPTinf performs against each detector we test.
| Detector | Score | Pass rate |
|---|---|---|
| Pangram | 61.8 | 62% |
| GPTZero | 67.4 | 67% |
| Originality.ai | 66.9 | 67% |
| Copyleaks | 65.2 | 65% |
| ZeroGPT | 66.7 | 67% |
At a glance
GPTinf sits 3.2 points behind SmartHumanizer and 5.4 points ahead of QuillBot in the October overall ranking. Meaning preservation at 67.4 is its strongest published component.
Detector results
GPTinf places ninth on all five detector-specific rankings. It scores 61.8 against Pangram, 67.4 against GPTZero, 66.9 against Originality.ai, 65.2 against Copyleaks and 66.7 against ZeroGPT.
The trade-off
The October table reports 65.6 for detector bypass and 58.1 for readability. All three published components rank ninth, so the result reflects a consistently lower position rather than one isolated weak metric.
Bottom line
GPTinf ranks near the bottom of the October field across every published measure. It finishes ahead of QuillBot overall, but the current data does not identify a detector or writing-quality advantage over the higher-ranked tools.
October 2026 results
The benchmark case for GPTinf
The current cycle places GPTinf ninth overall and ninth on each of the three published components.
Overall: 57.2
Ranked #9 of 10 tools, 5.4 points ahead of tenth place.
Detector bypass: 65.6
Ranked #9 among the 10 October tools for detector bypass.
Meaning: 67.4
Ranked #9 among the 10 October tools for meaning preservation.
Readability: 58.1
Ranked #9 among the 10 October tools for readability.
Related research
Explore Humanizer Arena research
See separate experiments and the site methodology for additional context.