StealthGPT
StealthGPT is an AI humanizer independently benchmarked by HumanizerBench. It ranked #12 of 12 in the August 2026 cycle with an overall score of 45.55/100 — our composite measure of how reliably it evades AI detectors while preserving the original meaning and readability.
Detector Bypass Rates
Percentage of samples classified as human-written by each detector.
Performance by Category
Bypass rate across writing contexts.
12-Month Trend
Overall score across the past year of testing cycles.
Session recordings
See the full methodology →Screen captures from the benchmark runs, kept for auditability. Hosted links may expire over time; the JSON in this repo is the source of truth.
Test markers
- 0:30stealthgpt · business_email · business_email
- 0:49stealthgpt · marketing_copy · landing_copy
- 1:11stealthgpt · application_essay · cover_letter
- 1:34stealthgpt · application_essay · personal_statement
- 2:18stealthgpt · blog_post · howto_blog
- 2:40stealthgpt · application_essay · cover_letter
- 3:00stealthgpt · discussion_board · discussion_post
- 3:21stealthgpt · blog_post · listicle_blog
- 3:55stealthgpt · academic_essay · lit_review
- 4:21stealthgpt · blog_post · howto_blog
- 4:46stealthgpt · academic_essay · argumentative_essay
- 5:09stealthgpt · marketing_copy · landing_copy
- 5:31stealthgpt · discussion_board · discussion_post
- 5:52stealthgpt · marketing_copy · landing_copy
- 6:15stealthgpt · application_essay · cover_letter
- 6:37stealthgpt · blog_post · listicle_blog
- 6:59stealthgpt · academic_essay · lit_review
- 7:31stealthgpt · application_essay · personal_statement
- 7:57stealthgpt · application_essay · personal_statement
- 9:39stealthgpt · academic_essay · argumentative_essay
- 10:18stealthgpt · marketing_copy · product_desc
- 11:00stealthgpt · news_article · news_article
- 11:24stealthgpt · discussion_board · discussion_post
- 11:45stealthgpt · academic_essay · argumentative_essay
- 12:08stealthgpt · marketing_copy · product_desc
- 12:26stealthgpt · business_email · business_email
- 12:40stealthgpt · news_article · news_article
- 12:55stealthgpt · business_email · business_email
- 13:19stealthgpt · news_article · news_article
- 13:39stealthgpt · marketing_copy · product_desc
- 14:01stealthgpt · blog_post · howto_blog
- 16:00stealthgpt · blog_post · listicle_blog
- 16:22stealthgpt · academic_essay · lit_review
Penalties applied this cycle
How penalties work →The overall score shown above already reflects these deductions.
- Meaning drift−1.0
The output's meaning drifted significantly from the original input.
- Length inflation ×2−2.0
The output ran much longer than the input, a common trick that pads text to dilute the AI signal.
- Length deflation ×2−2.0
The output came back much shorter than the input. The rewrite dropped content instead of paraphrasing it.
Minimum monthly pricing
- Monthly
- $25.50/mo
- Free tier
- No
How we tested
33 prompts across 7 writing categories, scored by 5 independent AI detectors. Every score is reproducible from the public methodology.