Pangram vs GPTZero: the 14 fields and the study results side by side
GPTZero carries 7 study records in the index and Pangram carries 0, the widest gap between the two cards on any field, by the count compiled on 8 November 2026 from the papers linked in the studies index. On every other field the two cards are read side by side below, with the vendor's wording and the page it came from, so the comparison rests on what each company publishes and what outside researchers measured.
Updated Fields checked 8 November 2026
Short answer
GPTZero has been tested in 7 published studies indexed here, with results from 9 false accusations in 18 human texts (Weber-Wulff 2023) to 0.00% false positives at threshold 0.5 (RAID 2024); no study in the index tested Pangram, whose false positive figure of 1 in 10,000 is the vendor's own (pangram.com). Pangram states 24 named languages and text, image and video detection; GPTZero states 5 languages and text plus an image tool (pangram.com, gptzero.me). Pangram's Individual plan is $20 per month; GPTZero's premium plans start at $12.99 per month per the vendor blog (pangram.com/pricing, gptzero.me). All read 8 November 2026. No detector score should be the only evidence in a misconduct or hiring decision.
How do Pangram and GPTZero compare on the 14 fields?
| Field | Pangram | GPTZero |
|---|---|---|
| 1. Name, vendor, official domain | Pangram, Pangram Labs, pangram.com | GPTZero, GPTZero LLC, gptzero.me |
| 2. Modalities detected | text, image, video | text, image (AI Vision, specs not published) |
| 3. Generators covered | OpenAI, Anthropic, Google, Meta, DeepSeek, xAI, Claude Opus 5, GPT 5.6, Gemini 3, Grok 4.3, Claude Sonnet 5, Claude Fable 5, GPT-5.4, GPT-OSS 120B, DeepSeek V4, Llama 3.3, Mistral Medium 3.5, Kimi K2.6, Qwen 3.7 Max, Gemma 4, Nemotron 3 Ultra, GLM 5.2, 26 models benchmarked | Claude, ChatGPT, GPT-3, GPT-4, GPT-5, GPT-6, Gemini, Llama, Deepseek, Grok, AI services based on those models |
| 4. Languages supported | 24 named: English, Arabic, Chinese, Czech, Dutch, French, German, Greek, Hindi, Hungarian, Italian, Japanese, Korean, Persian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Turkish, Ukrainian, Urdu, Vietnamese | 5: English, German, Portuguese, French, Spanish |
| 5. Input limits | Supports PDF, DOCX, RTF files and up to 100 files at a time. Minimum and maximum words per text check not published. Image: JPG, PNG, WebP, min 512x512px, up to 30 MB each. Video: MP4, MOV, WebM, min 720x720px, up to 500MB. API: 5 QPS realtime; bulk 1,000 billable units. | Widget: 10,000 characters per scan without an account. With a free account: texts up to 150,000 characters, or a batch of up to 250 files. File types and size not published. |
| 6. Free tier | 20 free credits daily (twenty free checks a day after signing up for the free plan, including AI assistance detection) | Free plan covering 10,000 words per month (vendor blog, Oct 25, 2026); 10,000 characters per scan without login |
| 7. Paid plans | Individual $20 / month; Professional $65 / month; Team $40 / month, 2 seats; API metered: Text (Pangram 4.1) $0.05 per 100 words, Image 1.0 $0.25 per image, Image Flash 1.0 $0.05 per image, Video $0.05 per second | Premium (individual plans): premium plans starting at $12.99/month; Professional Team Plan: 500,000 words per member, per month; min 2 seats (price not rendered) |
| 8. API | Yes, docs.pangram.com | Yes, docs.gptzero.me |
| 9. Integrations | Chrome extension, Firefox extension, Google Docs, Gmail, Google Classroom, Canvas, Moodle, Brightspace (D2L), MCP server | Chrome Extension (Gmail, Google Docs, Classroom, social media), Google Docs, Google Classroom, Canvas, Zapier, API |
| 10. Vendor accuracy and false positive claims | States accuracy of over 99% and third-party verification by researchers at the University of Chicago and The University of Maryland; "accurate on 99.6% of these AI-generated samples"; false positive rate 1 in 10,000 on an aggregate of public datasets containing tens of millions of written documents | States an accuracy rate of 99.9% when spotting AI-generated text vs. human writing, validated on RAID, detecting 95.7% of AI texts while incorrectly predicting 1% of human texts as AI; GPTZero 4o launch post, Oct 18, 2026: false positive rate of 1 in 10,402 on diverse student writing corpora |
| 11. Independent studies | 0 records | 7 records: Weber-Wulff 2023, Liang 2023, Walters 2023, Elkhatat 2023, Perkins 2024, RAID 2024, Chaka 2023 |
| 12. Data retention and privacy statement | Submitted text stored for as long as the account is active, deletable from the history tab; detection trained on a proprietary dataset that does not include user submissions; never shared or sold; zero data retention option for enterprise API | Privacy policy (May 20, 2026) lists submitted content as collected data; API inputs not stored; dashboard inputs stored and used in aggregate; FERPA, GDPR and CCPA compliance; deletion on request |
| 13. Launch year and ownership | 2023; Pangram Labs, co-founders Max Spero (CEO) and Bradley Emi (CTO), founded in Brooklyn, NY; no parent company stated | 2023; GPTZero LLC; announced June 23, 2026 that GPTZero plans to join Superhuman |
| 14. Rankings the card appears in | Languages supported | Independently tested, false positive rate, free tiers, languages supported, input limits |
Which studies tested both Pangram and GPTZero?
None of the nine studies indexed on the morning of 9 November 2026 tested Pangram, so the table below sets GPTZero's measured results beside Pangram's own statements, labelled as vendor statements, so the two columns are never read as the same kind of number (studies index, compiled 8 November 2026). Two records added later on 9 November 2026 tested both tools: Jabarian and Imas 2025 reports false positive rates of 0.0000 for Pangram and 0.0071 for GPTZero on 1,992 human passages at a 0.5 threshold, and false negative rates after StealthGPT rewrites of 0.0000 to 0.0500 for Pangram and 0.2552 to 0.7730 for GPTZero (NBER w34223, read 9 November 2026); Van Vlasselaer 2026 reports Pangram at 37 of 40 and GPTZero at 0 of 40 hybrid papers within their known AI range, with both tools correct on all 40 human papers (DOI 10.1007/s40979-026-00226-w, read 9 November 2026). The shared-studies count in the cards at a glance below reads those two records, and the three-way comparison with Originality.ai sets all of Jabarian and Imas 2025's figures for the three tools in one table.
| Source | Pangram | GPTZero |
|---|---|---|
| Vendor statement, pangram.com (read 8 November 2026) | False positive rate 1 in 10,000 on an aggregate of public datasets of tens of millions of documents; accurate on 99.6% of AI-generated samples | |
| Weber-Wulff 2023, 54 documents | 29 of 54 correct (54%, rank 8 of 14); 9 false accusations in 18 human-written documents; 3 of 36 AI documents undetected | |
| Liang 2023, 91 TOEFL and 88 US essays | One of seven detectors; average false positive rate across the seven 61.22% on TOEFL essays; per-detector rates in a chart only | |
| Walters 2023, 126 documents | GPTZero (Pro): 81% correct, 4% incorrect, 15% uncertain; 98% on GPT-3.5 papers, 57% on GPT-4; 2% false positive rate on 42 human papers | |
| Elkhatat 2023, 35 paragraphs | GPT-3.5 sensitivity 93% (14 of 15), specificity 80% (1 false positive in 5); GPT-4: 4 of 15 flagged | |
| Perkins 2024, 114 samples | Baseline mean accuracy 26.3%, lowest of seven; 1 false accusation in 10 human samples; 58 of 104 AI samples undetected | |
| RAID 2024, shared benchmark | False positive rate 0.03% at threshold 0.25 and 0.00% at 0.5; accuracy at a fixed 5% FPR of 98.8% on open-source chat models and 9.4% on non-chat models with sampling | |
| Chaka 2023, 21 chatbot responses and 30 translations | 8 of 21 English responses classified as AI-generated, 10 misclassified as human-written, 3 indecisive (rank 3 of 5); all 30 machine-translated responses classified as human-written |
Results as each study reports them: arXiv 2306.15666, arXiv 2304.02819, De Gruyter 10.1515/opis-2022-0158, Springer 10.1007/s40979-023-00140-5, arXiv 2403.19148, arXiv 2405.07940, DOI 10.37074/jalt.2023.6.2.12, all read 8 and 9 November 2026.
What do the two cost over 12 months?
| Plan | Monthly as stated | Annual as stated | Inclusions as stated | Checked on |
|---|---|---|---|---|
| Pangram Individual | $20 / month | Save $60, annually (12 x $20 less $60 is $180 by arithmetic) | Scan up to 300,000 words per month, 100 image detection scans monthly, similarity detection, Pangram MCP, Google Docs and Gmail integration, $10 in monthly API credits; 7-day free trial | 8 November 2026 |
| Pangram Professional | $65 / month | Save $240, annually ($540 by arithmetic) | 5x more words, 5x more image scans, $200 of API credits monthly | 8 November 2026 |
| Pangram Team | $40 / month, 2 seats | $20 / seat per month | 8 November 2026 | |
| Pangram API | Text (Pangram 4.1): $0.05 per 100 words; Image 1.0: $0.25 per image; Image Flash 1.0: $0.05 per image; Video: $0.05 per second | Bulk text API: 20% discount | 8 November 2026 | |
| GPTZero Premium (individual) | premium plans starting at $12.99/month | Individual plan cards did not render on 8 November 2026; the only readable figure is the vendor blog's starting price | 8 November 2026 | |
| GPTZero Professional Team Plan | Annual (Save 45%) / Monthly, price not rendered | 500,000 words per member, per month; min 2 seats | 8 November 2026 |
Prices in USD as shown with the $ symbol on pangram.com/pricing, gptzero.me/pricing and the GPTZero blog, checked 8 November 2026. GPTZero's pricing cards require JavaScript and did not render on that date; the card notes a browser recheck before any purchase decision.
One of three comparisons in the compare hub. The cards are Pangram and GPTZero, and the home board answers which AI detectors exist and who has tested them?
Which card fits which reader?
- Choose Pangram ifyour texts are in languages beyond the five GPTZero names: Pangram lists 24, including Arabic, Chinese, Hindi, Japanese and Korean (pangram.com, read 8 November 2026).
- Choose Pangram ifyou need image and video checks in the same account: both modalities are stated on the home page with file specifications (pangram.com, read 8 November 2026).
- Choose GPTZero ifyou want results that outside researchers have measured: 7 study records from 2023 to 2024, with a 0.00% false positive rate at threshold 0.5 in RAID 2024 and 9 false accusations in 18 texts in Weber-Wulff 2023 both on the record (arXiv 2405.07940, arXiv 2306.15666).
- Choose GPTZero ifyou check long documents free: 150,000 characters per text with a free account against Pangram's 20 free checks a day with no stated word count (gptzero.me/faq, pangram.com, read 8 November 2026).
- Choose Pangram ifAPI retention matters: a zero data retention option is stated for the enterprise API, while GPTZero states that API inputs are not stored but dashboard inputs are (pangram.com/data-privacy, gptzero.me/faq, read 8 November 2026).
How does the evidence compare, tool by tool?
Pangram
- Register ID
- AIDI-D-0006
- Study records
- 4 in the index: Jabarian and Imas 2025, Van Vlasselaer 2026, DAMAGE 2025, Epoch AI 2026
- Shared with GPTZero
- 4: Jabarian and Imas 2025, Van Vlasselaer 2026, DAMAGE 2025, Epoch AI 2026
- Vendor accuracy claim
- Pangram states accuracy of over 99% and third-party verification by researchers at the University of Chicago and The University of Maryland. We're accurate on 99.6% of these AI-generated samples.
- Free tier
- 20 free credits daily (twenty free checks a day after signing up for the free plan, including AI assistance detection)
- Languages
- 24 stated
GPTZero
- Register ID
- AIDI-D-0001
- Study records
- 16 in the index: Weber-Wulff 2023, Liang 2023, Walters 2023, Elkhatat 2023, Perkins 2024, RAID 2024, Chaka 2023, Krishna 2023, Orenstrakh 2023, Jabarian and Imas 2025, Van Vlasselaer 2026, Pegoraro 2023, Kar 2024, Alshammari and Rao 2025, DAMAGE 2025, Epoch AI 2026
- Shared with Pangram
- 4: Jabarian and Imas 2025, Van Vlasselaer 2026, DAMAGE 2025, Epoch AI 2026
- Vendor accuracy claim
- An accuracy rate of 99.9% when spotting AI-generated text vs. human writing, validated on RAID, detecting 95.7% of AI texts while only incorrectly predicting 1% of human texts as AI; mixed documents with a 96.5% accuracy rate. (FAQ)
- Free tier
- Free plan covering 10,000 words per month (vendor blog, Oct 25, 2026); 10,000 characters per scan without login
- Languages
- 5 stated
How this comparison is built
- Fields
- The 14 fields of every card, in card order
- Shared study records
- 4: Jabarian and Imas 2025, Van Vlasselaer 2026, DAMAGE 2025, Epoch AI 2026
- Fields checked
- 8 November 2026
- Records in the studies index
- 35 records
Both columns are drawn from the same 14 fields in the index data file, read on 8 November 2026 from the vendor page recorded on each card, and from the studies index, where each result is copied as the paper reports the figure, with the base stated. A vendor's accuracy or false positive statement is quoted and attributed and never placed in the same cell as a study figure. The rules, the recheck schedule (next pass 8 December 2026) and the correction route are on the methodology page; the independently tested ranking sorts every card on the study count used in the first sentence.
GPTZero is also set beside Sapling, a text detector with three shared records, in the Sapling vs GPTZero comparison, read 10 November 2026.
Where each figure comes from
Values as each vendor states them on pangram.com, pangram.com/pricing, pangram.com/data-privacy, pangram.com/about-us, gptzero.me, gptzero.me/faq, gptzero.me/pricing, gptzero.me/privacy-policy.html and gptzero.me/news/preserving-whats-human, all read 8 November 2026. An empty cell means the card has no published value for the field.
Questions about Pangram and GPTZero
Has any study tested Pangram?
None of the nine studies indexed on the morning of 9 November 2026 tested Pangram; two records added later that day, Jabarian and Imas 2025 and Van Vlasselaer 2026, tested both tools and are summarised above. Pangram's published false positive rate of 1 in 10,000 and its 99.6% figure on AI-generated samples are the vendor's own statements on pangram.com, read 8 November 2026; the study-sourced rate on the card is now 0.0000 of 1,992 (false positive rate ranking).
How many studies tested GPTZero?
Seven indexed records: Weber-Wulff 2023, Liang 2023, Walters 2023, Elkhatat 2023, Perkins 2024, RAID 2024 and Chaka 2023, the most of any card in the index (independently tested ranking, compiled 8 November 2026).
Which has the lower false positive rate?
The two figures are different kinds of number. Pangram states 1 in 10,000 on its own aggregate of public datasets (pangram.com); GPTZero recorded 0.00% at threshold 0.5 in RAID 2024, 2% of 42 human papers in Walters 2023 and 9 of 18 in Weber-Wulff 2023 (arXiv 2405.07940, De Gruyter, arXiv 2306.15666), all read 8 November 2026.
Which supports more languages?
Pangram names 24 languages against GPTZero's 5, as read on 8 November 2026 on pangram.com and gptzero.me. Pangram's list ends with "and more", so 24 is the named count.
Which costs less per month?
Pangram's Individual plan is $20 per month and GPTZero's premium plans start at $12.99 per month per the vendor blog of 25 October 2026 (pangram.com/pricing, gptzero.me, checked 8 November 2026). The two plans include different allowances, so the figures are not like for like.
Which has the larger free tier?
GPTZero states 10,000 words per month and 150,000 characters per text with a free account; Pangram states 20 free checks a day with no word count (gptzero.me, gptzero.me/faq, pangram.com, read 8 November 2026). The index does not convert checks per day into a monthly figure.
Do both integrate with Canvas and Google Classroom?
Yes: Pangram names Google Classroom, Canvas, Moodle and Brightspace (D2L); GPTZero names Google Classroom and Canvas, both read 8 November 2026 (pangram.com, gptzero.me).
Sources and revisions
Sources
- Pangrampangram.com
- Pangram pricingpangram.com/pricing
- Pangram data privacypangram.com/data-privacy
- Pangram aboutpangram.com/about-us
- GPTZerogptzero.me
- GPTZero FAQgptzero.me/faq
- GPTZero pricinggptzero.me/pricing
- GPTZero blog, 25 October 2026gptzero.me/news/best-ai-detectors
- GPTZero privacy policygptzero.me/privacy-policy.html
- Weber-Wulff et al. 2023arxiv.org/pdf/2306.15666
- Liang et al. 2023arxiv.org/pdf/2304.02819v3
- Walters 2023degruyterbrill.com/document/doi/10.1515/o...
- Elkhatat et al. 2023link.springer.com/content/pdf/10.1007/s40...
- Perkins et al. 2024arxiv.org/pdf/2403.19148
- Dugan et al. 2024 (RAID)arxiv.org/pdf/2405.07940
- Chaka 2023doi.org/10.37074/jalt.2023.6.2.12
