GPTZero AI score
The saved detector result for the humanized text. The main ranking gives each tested source model equal influence. A lower score does not prove that a person wrote the text.
AI detection and stylistic analysis
AI Humanizer benchmark
Which AI humanizers actually change the writing? Compare their GPTZero results and writing style across Claude Fable 5.1, GPT-6 Astra, GPT-4, and Gemini 3.1 Pro. Inspect the scores, then read the original and humanized text for yourself.
Compare the original LLMs in the Stylistic LLM Writing benchmark
Detailed Humanizer comparison
Lower GPTZero scores rank higher. AI Style score measures writing patterns separately. Missing scans are not treated as zero.
The latest Humanizer table is being prepared. Please check back shortly.
| Rank | Humanizer / model |
|---|
What the scores tell you
AI Style score
A separate comparison of sentence lengths, noun contextualization, heavy grammar, personal tone, rhetorical templates, overused vocabulary, punctuation, section variation, and GPT-2 predictability.
Original versus humanized
Choose a source model to compare each Humanizer with the original. Open a row to inspect its individual writing tasks and read both versions.
Coverage matters
Not every Humanizer has the same number of completed tasks or detector scans. Check the coverage before comparing close scores. Results describe these saved tests, not every text a tool might produce.
AI Humanizer benchmark FAQ
How are Humanizers ranked?
By their average GPTZero AI score across tested source models, lower first. Each available source model has equal influence. AI Style score breaks ties. Humanizers without scans appear after scanned results.
Were the original models scanned by GPTZero?
No. Original LLM rows use a 100% reference default and the human baseline uses 0%. These are clearly labeled defaults, not measured detector results. GPTZero scans apply to humanized outputs.
Does a lower score mean better writing?
Not necessarily. Detector scores and style measurements do not establish factual accuracy, usefulness, originality, or reader preference. Read the paired outputs to check whether the rewrite preserved the meaning.
Why compare different source models?
A Humanizer can perform differently on each model's style. The source-model tab separates these effects instead of hiding them in a single average.
How do the results stay up to date?
This page uses the benchmark owner's server-published results, including the same metric labels as the app. New tests appear after the updated table is synchronized. No tool is tested or charged when you view this page.
Can an AI humanizer guarantee bypassing detectors?
No. A result from one detector on one set of samples is not a guarantee for other texts, detector versions, or contexts.