Pangram vs GPTZero
Pangram is the better detector for checking someone else's essay: it states the lowest false positive rate of the two and held up best in a 2025 independent test. False alarms on human writing in run RUN-2026-11-A: Pangram Test in progress, GPTZero 0 of 100.
Tested: In progress since 9 November 2026. Page updated: .
Short answer
Verdict: Pangram. It states the lowest false positive rate of the two and held up best in a 2025 independent test.
Run RUN-2026-11-A started on 9 November 2026 and is in progress: GPTZero and Originality.ai are scored on the first 120 samples, and every other tool shows a test in progress tag until its readings are in.
GPTZero has the more generous free plan and sentence highlights anyone can use. Treat any flag from either tool as a question, not an answer.
See them side by sideBoth readings on one essayDownload the raw files
Side by side
The same fields for both tools, from the same run.
| Field | Edge | ||
|---|---|---|---|
| Access | Web, LMS integrations and API | Web, browser extension and API | Even |
| Price | $20 a monthChecked on | CA$25.99 a monthChecked on | Even |
| False alarms on human writing | Test in progress | 0 of 100 | Test in progress |
| Detection on AI text | Test in progress | 20 of 20 | Test in progress |
| Report detail | Window scores | Sentence highlights | Pangram |
Which is more accurate?
Accuracy here means both halves: AI text caught, and human writing left alone.
Pangram states a false positive rate of 1 in 10,000; GPTZero says its rate is under 1% on its developer page. In a 2025 University of Chicago working paper testing Pangram, Originality.ai, GPTZero and an open model on matched human and AI passages, only Pangram met a false positive cap of 0.5%. That paper is a preprint, not peer reviewed. For the free head-to-head, see GPTZero vs ZeroGPT, and for institutions, GPTZero vs Turnitin.
One student essay, the two tools scored so far
Sample H-002 · student essay, written without AI · run RUN-2026-11-A · Pangram and GPTZero readings are added when the run scores both
GZGPTZero
OAOriginality.aiWhile the idea of having a fully-autonomous is very real, scientists are still very far from it. Research is being conducted every day to improve the efficeincy and safefty of these new semi-autonomous vehicles. In my opinion, driverless cars are inevitable and I would not doubt that I will end up seeing one in my lifetime. But as of right now, cars can't do as much as scientists are hoping for in the future. With that being said, I will there will be a more positive outcome of driverless cars than negative.
Ch. 10% AICh. 22% AIMany skeptics say dirverless cars will never be possible. But, what they don't realize is that we are closer than we ever have been before regardless of how much research is still necessary to succesfully make a fully autonomous car. Google's concept car has been able to drive over half a million miles without a crash. Their car still requires a driver, but it is only for more advanced tasks such as traffic issues or roadwork. Other companies have also attempted to create a driverless concept car and the concepts have showed potential. They are far from driverless but they are going in the right direction. BMW announced that their car could handle driving speeds of up 25 mph but still require the focus of a human being behind the wheel.
Ch. 10% AICh. 22% AIWhile I am very supportive of driverless cars, I think there will always be flaws, no matter the size. I think there will always be malfunctions but that does not mean that those can be fixed either. The research that has been put into these vehicles have been very strenuous and have proven that there will never not be flaws in creating an autonomous car. They still may crash but the probability of a fatal crash in a driverless is much lower than the probability of a fatal crash due to human error. Humans get distracted thus creating more chances for an accident to occur. With driverless cars, there is no chance of human error or the car getting distracted because driving is what it was built for.
Ch. 10% AICh. 22% AIThere are not many current laws or regulations on autonomous cars, so even if Google or BMW were to finish research and create a flawless and driverless car, they wouldn't be able to manufacture them on a commercial level. This is because there wouldn't be any laws covering liability in the case of an accident, if an accident were to somehow occur. It is undetermined who would be at fault for the accident according to the state. I think the company should be more at fault, considering they are advertising these cars as "driverless".
Ch. 10% AICh. 22% AIWith my opinion on driverless cars being said, I still think they are a good idea and will benefit society in the long run and keep more people safe. There are bumps in the road regarding laws and the time it will take to get to where big name companies want to be. But, if we are all patient and let scientists conduct quality research, I see a bright future for fully autonomous cars.
Ch. 10% AICh. 23% AIScroll the chart sideways to read every paragraph.
| Subset | Pangram | GPTZero |
|---|---|---|
| Student essays (60) | Test in progress | 0 of 60 |
| Non-native English essays (20) | Test in progress | 0 of 20 |
| Professional prose (20) | Test in progress | 0 of 20 |
| Raw AI text | Test in progress | 20 of 20 |
| Human-edited AI text (20) | Test in progress | Test in progress |
Access and price
What you can do without paying, what paying adds, and when we checked.
PGPangram
- Free plan
- Yes, after sign-up
- Free word limit
- 20 checks a day
- Paid plan
- $20 a month
- Words per scan, paid
- 300,000 words a month
- Batch upload
- Paid plans
- School plans
- Yes, through sales
Checked on
GZGPTZero
- Free plan
- Yes, with an account
- Free word limit
- 10,000 words a month
- Paid plan
- CA$25.99 a month
- Words per scan, paid
- 300,000 words a month (Premium)
- Batch upload
- Up to 3 files on the free plan
- School plans
- Yes
Checked on
Reports
A score you cannot inspect is hard to act on. What matters is whether the report shows which sentences drove the number, so you can read them yourself.
GPTZero scores each paragraph and highlights sentences. ZeroGPT gives one percentage for the whole text with highlights, which makes a single formal paragraph harder to spot.
| Report shows | Pangram | GPTZero |
|---|---|---|
| Report style | Window scores | Sentence highlights |
| Who can sign up | Anyone | Anyone |
Which should you use?
The same two tools, a different answer for each reader.
A flag that will be argued over
PGPangramThe careful second reading.
Teachers on a budget
GZGPTZeroA free plan with highlights.
Developers
PGPangramA per-word API price.
Students
GZGPTZeroSee which sentences look predictable.
Three alternatives
If neither fits, these are the next detectors to try, from the same run.
No detector score should be the only evidence in a misconduct case.
Questions
Is Pangram more accurate than GPTZero?
On published evidence, yes: the 2025 University of Chicago working paper found Pangram alone met a 0.5% false positive cap. In run RUN-2026-11-A so far, GPTZero flagged 0 of 100 human texts as AI; Pangram's readings are in progress.
Is Pangram free?
Pangram has a free plan with 20 checks a day after sign-up, per its own page on 8 November 2026.
Which has an API?
Both. Pangram lists $0.05 per 100 words; GPTZero sells API plans separately.
Related reading
Most Accurate AI Detector (2026): Accuracy RankedSorted only on false alarms and AI caught.
GPTZero vs ZeroGPT: Which AI Detector Is Better?Verdict for each type of user.
GPTZero Alternatives (2026): Picks by Reason to SwitchPicked by the reason you are switching.
AI Detector Comparisons: Head-to-Head ResultsTwo detectors, the same samples. Each comparison sets both readings over one.