NoteGPT
NoteGPT ranked #13 of 13 in the July 2026 cycle with an overall score of 44.04/100. ZeroGPT passed its output most often (64.6%), and GPTZero and Copyleaks passed none of it. It did best on business emails and worst on academic essays.
NoteGPT's AI humanizer is one tool in the NoteGPT platform, which also offers AI note taking and video and PDF summaries. It has nine preset tones plus a custom tone, with settings for output length and similarity to the original. Its standalone humanizer is free without signing up; paid NoteGPT memberships cover other platform features.
- Overall score
- 44.04/100
- Detector bypass
- 0.0%
- Overall rank
- #13of 13
- Prompts tested
- 33
Detector Bypass Rates
Each detector's average human-likelihood score on this tool's outputs, where 100 means every output was scored as human.
Performance by Category
Category score across writing contexts: bypass credit only for real rewrites, blended with each category's meaning and readability. How it's computed
Strengths
- Readability of 74.5, above the field median of 60.3
- No scoring penalties this cycle
Weaknesses
- Bypass rate of 0.0%, below the field median of 73.4%
- GPTZero and Copyleaks classified none of its runs as human-written
- Weakest writing category: academic essays (33.9)
From the July 2026 cycle, compared with the other 12 tools ranked in that cycle.
Evidence: every run
Every prompt NoteGPT was run on in July 2026, and how each of the 5 detectors scored the output. Green: the detector called it human-written. Red: it was caught.
0 of 33 outputs passed all 5 detectors; 33 were caught by at least one.
| Prompt | GPT GPTZero | Win Winston AI | Zero ZeroGPT | Copy Copyleaks | Orig Originality.ai | Median | Recording |
|---|---|---|---|---|---|---|---|
| Argumentative Essay · GPT-5.5 | caught | caught | caught | caught | caught | 0.00 | |
| Listicle Blog · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Cover Letter · Gemini 3.5 Flash | caught | caught | caught | caught | caught | 0.00 | |
| Business Email · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Howto Blog · Gemini 3.5 Flash | caught | caught | caught | caught | caught | 0.00 | |
| Lit Review · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Product Desc · GPT-5.5 | caught | caught | caught | caught | caught | 0.00 | |
| Landing Copy · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Personal Statement · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Business Email · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Product Desc · Claude Sonnet 5 | caught | caught | caught | caught | caught | 0.00 | |
| Listicle Blog · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Product Desc · Gemini 3.5 Flash | caught | caught | caught | caught | caught | 0.00 | |
| Howto Blog · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Cover Letter · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Argumentative Essay · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Discussion Post · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Lit Review · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Cover Letter · Claude Sonnet 5 | caught | caught | caught | caught | caught | 0.00 | |
| Business Email · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Discussion Post · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Personal Statement · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Landing Copy · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| News Article · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| Listicle Blog · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Landing Copy · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Discussion Post · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Argumentative Essay · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Lit Review · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Howto Blog · GPT-5.5 | caught | caught | caught | caught | caught | 0.00 | |
| News Article · GPT-5.5 | caught | caught | passed | caught | caught | 0.00 | |
| Personal Statement · Gemini 3.5 Flash | caught | caught | passed | caught | caught | 0.00 | |
| News Article · Claude Sonnet 5 | caught | caught | passed | caught | caught | 0.00 | |
| Detector rate (mean human-likelihood) | 0.0 hardest | 0.2 | 64.6 | 0.0 | 0.1 | ||
| Runs passed | 0/33 | 0/33 | 25/33 | 0/33 | 0/33 |
A dot is green when the detector's human-likelihood score for that run was 0.50 or higher. The detector rate is the mean of that detector's scores across the runs, the same number the detector pages rank by; it is not the count of green dots, which the Runs passed row shows. Hover a dot for the score.
How to read these scores
Each detector returns a human-likelihood on a common 0 to 1 scale, where 1 means it judged the text human-written and 0 means it flagged it as AI. On these pages a verdict counts as passed when that score is at least 0.50, the midpoint of the detector's own scale. That threshold exists only to draw the chips: the bypass rate on the leaderboard is the mean of each test's median score across the 5 detectors, a continuous number, so a tool's pass count and its bypass rate will not be the same figure.
History
Overall score compared with other ranked tools.
19.8 points below the median of 12 other tools in July 2026.
The shaded band is the 25th–75th percentile of other tools in each cycle. Scores are out of 100; higher is better.
Monthly pricing
Standard USD rates, billed month to month. Annual discounts and promotions excluded.
The standalone humanizer is free; paid NoteGPT memberships cover other platform features.
Free to use
Free access and trials
The standalone humanizer is advertised as free, with no signup and unlimited use.
- The live input counter allows 20,000 characters, while the same page’s prose says 10,000. The per-request limit is therefore inconsistent across official information.
- Paid platform quotas are not humanizer quotas. Standard undiscounted platform rates could not be resolved from conflicting reference prices.
Sample output
One AI-written passage from the July 2026 cycle and what NoteGPT returned for it. This prompt was picked by a fixed rule, not by how the tool scored.
AI-written input
Every tool on this prompt →Listicle Blog · Blog Post · written by Gemini 3.5 Flash · 405 words
**10 Tips for running your first 10k race**
So, you’ve signed up for your first 10k? Heck yeah! Double the distance of a 5k might feel a little daunting, but trust me, you’ve got this. Crossing that finish line is one of the best feelings in the world.
To help you transition from "Oh no, what did I do?" to crushing your race day, here are 10 super casual, highly practical tips for your first 10k.
### 1. Don't buy brand-new shoes the week of the race Seriously, don't do it. Race day is not the time to break in new kicks. Stick to the trusty trainers you’ve been training in to avoid blister city.
### 2. Follow a plan (and actually stick to it) Don't just wing it. Find a simple 8-to-10-week 10k training plan. Consistency is your secret weapon here, not speed.
### 3. Pace yourself (seriously, slow down!) When the starter pistol goes off, the adrenaline will tempt you to sprint. *Resist.* Start slower than you think you need to. You’ll thank yourself at kilometer 8.
### 4. Practice your pre-race breakfast Nothing new on race day! Figure out what fuel works for your stomach during training runs. A banana with peanut butter or a piece of toast is usually a safe bet.
### 5. Walk breaks are not cheating If you need to walk through the water stations or up a steep hill, do it. It’s about finishing, not being a hero.
### 6. Do a dress rehearsal A week before, go for a run in the exact outfit you plan to wear on race day. Make sure nothing chafes, rides up, or drives you crazy.
### 7. Hydrate the week *before* Gulping a gallon of water the morning of the race will only lead to sloshing and emergency bathroom breaks. Hydrate steadily in the days leading up to the event.
### 8. Study the course map Knowing where the hills are (and where the finish line is) helps you mentally chunk the race into manageable parts.
### 9. Make an epic playlist When the legs get heavy, a high-energy beat can literally pull you forward. Curate your favorite hype tracks.
### 10. Smile for the cameras! You’re doing something amazing. When you see the race photographers, chin up, shoulders back, and flash a grin.
Have fun out there, enjoy the crowd energy, and get ready to rock that finisher's medal!
Show the full input Show less
NoteGPT output
392 words (input 405)
10 Tips for Crushing Your First 10k
Signed up for your first 10k? Awesome. Sure, it sounds twice as tough as a 5k, and maybe you're wondering what you got yourself into—but you CAN do this. That finish line? Nothing beats it.
To replace your nerves with confidence, here’s a quick list of what actually helps when you’re gearing up for race day:
1. Stick with your usual shoes Don’t risk shiny new sneakers right before the race. Wear the ones you’ve been training with, trust me—you don’t want blisters as your race day souvenir.
2. Choose a training plan…and don’t bail on it Random runs won’t get you there. Find a solid 10k training plan, eight to ten weeks is plenty. Don’t obsess over speed; showing up for your workouts makes all the difference.
3. Start off slow—seriously, slower than you think Race mornings are electric. The temptation to blast off with everyone else is real, but hold back. If you pace yourself at the start, you’ll still have gas left near the end.
4. Test your race morning breakfast ahead of time Don’t try anything wild on the big day. Use your training runs to figure out what works—maybe it’s a banana with peanut butter, or toast. Simple is safe.
5. Take walk breaks if you want to Walking through water stations or up a tough hill isn’t cheating. The goal is to finish, not to become a superhero mid-race.
6. Do a full dress rehearsal A week before the race, run in your race day clothes. Make sure nothing rubs you raw or annoys you. It’ll save you drama.
7. Hydrate early, not just on race morning Don’t chug water right before the start—that’s a recipe for bathroom runs. Drink consistently during the week leading up.
8. Know the course layout Check out the course map. Knowing where hills and the finish line are helps you mentally break up the race.
9. Make a playlist that pumps you up Great songs really do help when you’re tired. Build a playlist that gets you moving.
10. Smile for the cameras! You’re out there doing something awesome. When you spot a photographer, don’t forget to smile. You’ll want those pics later.
Most important: soak up the excitement, enjoy the crowd, and get ready to show off that medal. You’ve got this.
Show the full output Show less
- GPTZero0.00caught
- Winston AI0.00caught
- ZeroGPT0.72passed
- Copyleaks0.00caught
- Originality.ai0.00caught
Passed 1 of 5 detectors.
Every NoteGPT run in July 2026 →Each chip is one detector's score for that output, from 0 (AI) to 1 (human). A score of 0.50 or higher counts as a pass and shows green; below that the detector caught it and the chip shows red.
Compare with
Questions about NoteGPT
Answers are generated from the benchmark data and update every month. Last updated with the October 2026 cycle.
Is NoteGPT good?
NoteGPT ranked #13 of 13 in the July 2026 HumanizerBench cycle with an overall score of 44.04/100, below the field median of 62.84. On the parts of the score, it placed tied for #12 of 13 on detector bypass, #9 of 13 on meaning preservation and #2 of 13 on readability.
Does NoteGPT bypass AI detectors?
In the July 2026 cycle, NoteGPT had a detector bypass rate of 0.0% (tied for #12 of 13 tools). 0 of its 33 outputs passed all 5 detectors, and 33 were caught by at least one. Its average human-likelihood score from each detector, where 100 means every output was scored as human: ZeroGPT 64.6, Winston AI 0.2, Originality.ai 0.1, GPTZero 0.0 and Copyleaks 0.0.
Can NoteGPT bypass GPTZero?
In the July 2026 cycle, NoteGPT's average human-likelihood score from GPTZero was 0.0 out of 100 (tied for #12 of 13 tools against GPTZero). The highest average GPTZero score that cycle was 94.1, by Undetectable.ai. See every tool's GPTZero results.
Is NoteGPT free?
The standalone humanizer is advertised as free, with no signup and unlimited use.
Is NoteGPT legit?
NoteGPT is an AI humanizer at notegpt.io. HumanizerBench last tried it in the August 2026 cycle, when it was unavailable during testing, so it was not ranked. In that cycle, all 33 of its runs were site or access failures. It was ranked in 2 earlier monthly cycles, most recently July 2026, when it completed 33 of 33 test runs. Every input and output from its runs is published with each detector's verdict.
Does NoteGPT keep the original meaning?
NoteGPT scored 69.1 out of 100 for meaning preservation in the July 2026 cycle (#9 of 13; field median 72.9). The score measures how close each rewrite stays to the meaning of the original text. Its readability score was 74.5 (#2 of 13). It drew no scoring penalties that cycle.
Which kinds of writing does NoteGPT handle best?
In the July 2026 cycle, NoteGPT did best on business emails (45.9/100) and worst on academic essays (33.9/100). On academic essays it ranked #13 of 13.
How has NoteGPT ranked over time?
NoteGPT placed #13 in both HumanizerBench cycles it was ranked in, starting with June 2026. Its overall score went from 45.34 in June 2026 to 44.04 in July 2026. It has not been ranked since July 2026. When HumanizerBench last tried it, in August 2026, the tool was unavailable during testing.
What are the best alternatives to NoteGPT?
NoteGPT was not ranked in the October 2026 cycle. The top three that cycle were WriteHuman (#1, 74.89/100), StealthGPT (#2, 74.50/100) and Undetectable.ai (#3, 71.18/100).