Turnitin AI Checker Benchmark: Testing Turnitin0 Against Reilaa, Scribbr, Quetext, Originality.ai, Zerogpt and More

A benchmark is only as credible as its method. This one compares seven tools — Turnitin0, Reilaa, T-detector (turnitindetector.com), Scribbr, Quetext, TurnitinEye, Originality.ai, ZeroGPT, and AIDetectPlus — across five dimensions that actually matter to a student facing a submission deadline:

1. Report authenticity — does the tool produce a genuine Turnitin AI detection report and similarity report, or an independent estimate?
2. Repository behaviour — is the uploaded file added to a student paper database, or checked non-repository?
3. Detection evidence — are there published, word-level accuracy studies behind the claims?
4. Turnaround and file handling — how fast, what formats, what size and word limits?
5. Independent verification — is there user feedback outside the vendor's own homepage?

Where a vendor only publishes claims on its own site, we label those as vendor claims. Where Turnitin0 has published research reports with word-level accuracy figures, we cite those figures directly. We also rely on Turnitin's own documentation and faculty guidance to explain what the AI writing indicator actually means, because a large part of this benchmark is about understanding the target you are trying to hit [1][2].

Before comparing tools, it helps to be precise about the output. Turnitin's AI writing detection appears as a blue AI indicator alongside the familiar similarity report. When the detector is confident, it shows a percentage; when detection falls below its 20% confidence threshold, it displays an asterisk (*%) instead of an exact number, and a dash (--) means nothing was identified [2].

That asterisk is not a rounding quirk — it is a signal that the score is less reliable at that level. This matters enormously for benchmarking, because two tools can both "pass" while producing very different underlying evidence. It also explains why Turnitin0's own reporting shows *% rather than an exact figure when AI detection sits below that threshold. That is a transparency point, not a defect: it mirrors exactly what an instructor sees.

Turnitin's AI detector is also not available to students directly — it is an instructor-side feature, and students typically do not see the report unless the instructor releases it [2]. That asymmetry is the entire reason a pre-submission Turnitin check service exists.

Turnitin0 is an independent service and is not affiliated with Turnitin, LLC. It offers two products: a Turnitin checking service and an AI humanizer.

The Turnitin checking service accepts.docx,.pdf, or.txt files, English only, with a word count greater than 300 and less than 30,000, and a file size under 20 MB. Each order includes two downloadable PDFs in a single checkout: a Turnitin AI detection report and a similarity/plagiarism report. Critically, the reports are identical to what professors see in their LMS.

The non-repository guarantee is the headline feature. Files are checked without being added to Turnitin's student paper database, and reports are not shared with third-party databases. Users can delete files from their account, and there is no subscription.

Turnaround is fast: under 15 minutes in 98% of cases, most orders finishing within 5–15 minutes, with an average under 15 minutes and rare queue spikes guaranteed within 30 minutes.

The AI humanizer supports.docx or.txt, English only, under 90 MB, and is built for text drafted with ChatGPT, Claude, or Gemini. It preserves meaning, citations, headings, and.docx formatting, and carries a score promise: lower the Turnitin AI score to *% or below 20%, or even 0%, or a full refund. Notably, 98.2% of humanizer orders are re-checked with Turnitin.

Scale and satisfaction: 100,000+ Turnitin AI and similarity reports delivered, 20,000+ students worldwide across the United States, United Kingdom, Canada, Australia, New Zealand, and Ireland, and a 4.9/5.0 satisfaction rating.

What Turnitin0's Own Research Shows

Turnitin0 has published seven research reports, and they are unusually candid — including results that are unflattering to the humanizer.

Two of these findings deserve emphasis. First, Turnitin is extremely good at catching raw LLM output — above 97% word-level accuracy for GPT-5.6-Sol and above 99% for Claude Fable-5. Second, Turnitin is much weaker against AI-polished human writing, where accuracy dropped to 47.54% overall and hit 0% in some majors. And the humanizer's 76.44% overall evasion rate is honest reporting, not a victory lap: Education hit 100%, but English sat at just 55.41%.

That is the kind of disclosure a benchmark should reward. If you want to see the full methodology behind these numbers, the Turnitin AI detector research library is the place to start.

Verified Customer Feedback

Turnitin0's user reviews are specific and attributable, which is more than most competitors in this comparison can offer.

"Process easy, fast, efficient; report came back much faster than expected." — Raini Dipré (CA), 5 stars

"Report complete after about 20 minutes; can download AI and similarity score reports at the same time." — may zin (SG), 4 stars

"Used Turnitin0 several times; reports delivered quickly; Humanize helpful when revising." — daniela pellegrini (GB), 5 stars

"Easy, quick, helpful for checking and improving academic writing; liked Humanize for sounding more natural while keeping original meaning." — Shubham Pachauri (IN), 5 stars

"Best site for turnitin scans, authentic and simple to use." — Encrypted (GB), 5 stars

"Great Service, 100% Legit and works." — Taksh Patel (AU), 5 stars

Other reviewers include b c (US), who used the site for assignments, plagiarism checking, and awareness of AI, and Shawn Thakur (AU), who found it easy to use and on time.

Honest Limitations

A credible benchmark states the downsides. Turnitin0's AI score display shows *% instead of an exact percentage when AI detection is below its 20% confidence threshold. There is no free word quota or free trial for the humanizer. The Trustpilot profile carries a TrustScore of 4.3/5 from 9 reviews, all in the last 12 months, and the company has not recently invited customers, so those reviews may not be representative. The checking service is English-only, requires more than 300 and fewer than 30,000 words, and caps files at 20 MB; the humanizer caps files at 90 MB.

Reilaa: Free, Unlimited, and Unverified

Reilaa positions itself as a free AI detector for students, offering unlimited scans with no sign-up required and optional Turnitin reports. It claims essays are immediately deleted, that it does not sell user data, and that detection takes under 5 seconds.

Those are attractive claims, and the underlying problem Reilaa addresses is real. Its homepage cites Reddit cases where students received a 62% AI score on an original essay and faced a Student Accountability meeting, were flagged and investigated for academic dishonesty, or were accused of using AI twice despite not using it, resulting in a Conduct Hearing.

The weakness is verification. No independent user feedback was provided, and every performance figure — 0 essays stored, 0 emails collected, unlimited free scans, 100% free forever, under 5 seconds detection — is a vendor claim from the homepage. Reilaa is a reasonable first filter, but it is not a substitute for a real Turnitin report.

T-detector (turnitindetector.com): Similar Promise, Narrower Input

T-detector is a third-party service claiming to provide Turnitin AI and plagiarism reports without storing documents in Turnitin's database. It states that users receive a Similarity Report and an AI Report, that supported formats are.pdf and.docx only, that documents must contain 320–29,999 words, and that processing usually takes 5–20 minutes.

Its privacy posture is detailed: files are never submitted to Turnitin's database or stored anywhere else, users can delete data anytime, and if they forget, the system automatically erases it after 24 hours. It claims top-tier encryption, real-time status updates on a History page, email delivery of reports, and that reports include both an AI-generated text score and an AI-paraphrased score with flagged sentences highlighted.

The gaps: no.txt support, a narrower word range than Turnitin0, and no independent user feedback to verify any of the claims. It is a plausible alternative, but the evidence base is thinner.

Scribbr: Academic Services, Not a Turnitin Check

Scribbr is a broad academic support platform rather than a Turnitin checker. It offers proofreading, plagiarism checking, citation generation, and AI detection, with human proofreading claimed at a 3-hour turnaround and AI proofreading claimed to improve an entire paper within 5 minutes. Its citation generator supports APA, MLA, Chicago, AMA, IEEE, and ACS, and the site is available in English, Dutch, German, French, Italian, Spanish, Swedish, Danish, Finnish, and Norwegian Bokmål.

As a writing-improvement ecosystem, Scribbr is genuinely strong, and the multilingual footprint is unusual. But it does not deliver a Turnitin AI detection report or a Turnitin similarity report, so it cannot answer the question this benchmark is asking.

Quetext: Broad Toolkit, Vendor-Only Evidence

Quetext combines a plagiarism checker and AI detector with a wider writing suite. It claims DeepSearch™ technology searching billions of sources, ColorGrade™ feedback distinguishing exact, near-exact, and fuzzy matches, an interactive snippet text viewer, comprehensive plagiarism and AI scoring, bulk scan, a Chrome extension, and a developer API with RESTful documentation. Its toolset includes an AI Humanizer, AI Summarizer, AITutorMe paraphrasing tool, citation generator, and grammar checker. The homepage states DeepSearch version 2.0 and claims over 10 million students, teachers, and professionals helped.

The limitation is the same as with several others here: every performance, accuracy, and privacy statement is a vendor claim, with no independent user feedback provided to verify it.

TurnitinEye: Claims Turnitin Integration Without Evidence

TurnitinEye markets itself as a "#1 Turnitin Alternative," claiming advanced neural networks for AI detection, cross-referencing against billions of web pages and academic papers, instant PDF reports with institutional formatting, source attribution with real-time verification and links, and support for PDF, DOCX, TXT, and direct text input up to 10MB. It claims documents are never stored or used for training, that it is trusted by 10,000+ users and universities worldwide, and — most significantly — that it has direct integration with Turnitin to provide authentic originality reports.

That last claim is the problem. No evidence is provided for the claimed Turnitin integration, no independent user reviews are available, and no detection accuracy or false-positive rates are disclosed. The sample report on the homepage shows a 12% Similarity Index, 04% AI Probability, Similarity Check v3.2, and reference 829-VT-SUB — but a sample is not a validation study.

Originality.ai: The Strongest Independent Research Story

Originality.ai is the most credible non-Turnitin detector in this comparison. It claims proven accuracy in multiple peer-reviewed third-party studies, supports detection across leading LLMs including GPT-6 Astra (Sol, Luna, Terra), Claude Fable 5, Gemini 3, Kimi K3, and DeepSeek V4 (Flash & Pro), and offers 3 free AI scans per day up to 2,000 words each. It is trained on adversarial data to detect AI paraphrasing tools like Quillbot or Grammarly, and integrates via Chrome extension, Google Docs, Firefox, Moodle plugin, and API. Additional tools include a plagiarism checker, grammar checker, readability checker, fact and AI hallucination checker, content quality score, guideline checker, and deep scan.

If your goal is to understand whether text reads as AI-generated, Originality.ai is excellent. If your goal is to know what your professor's Turnitin screen will show, it is still an estimate — a different model with different training data.

ZeroGPT: Wide Tool Suite, Vendor Claims Only

ZeroGPT markets itself as a trusted AI detector for ChatGPT, GPT-6, and Gemini. It claims a high-accuracy model trained on all languages and multiple LLM models, support for all languages, highlighted sentences with a gauge showing the percentage of AI in text, automatically generated.pdf reports, batch file upload, and millions of users. Its tool suite is the broadest here: AI/GPT detector, AI humanizer, AI image detector, AI video detector, plagiarism checker, AI paraphraser, AI grammar check, AI summarizer, AI translator, word counter, dictionary, and AI email helper.

Operationally, the free checker accepts up to 15,000 characters, and the homepage lists English, French, German, Spanish, Italian, and Portuguese. Every pro, however, is a vendor claim with no independent verification.

AIDetectPlus: Explanation-First Detection

AIDetectPlus bundles AI detection, AI humanization, and plagiarism checking, sold via non-expiring credits. Its differentiator is explanation: the detector marks sentences it believes are AI-written and states why each was flagged — for example, repetitive rhythm or stock hedging phrases — and the humanizer shows line-by-line changes. It supports.docx,.pdf,.txt,.html, and pasted text, claims compatibility with ChatGPT, Gemini, Claude, and any GPT model, and offers free AI detection without an account. Credits are purchased once, do not expire, and are shared across all three tools, with an Enterprise plan offering bulk detection and humanization, API access, exportable reports, volume discounts, and a dedicated account manager.

The vendor reports a 4.8 TrustPilot rating, 100k+ writers, 1.2M documents analyzed, and a Product Hunt #2 Product of the Day. Testimonials include Owdy (@Hadeel_Naily), who says AIDetectPlus flagged AI content better than GPTZero or Copyleaks and values non-expiring credits over a subscription, and Kijan (@kijanayamwingi), who reports a 98% human score from GPTZero after humanizing. The honest caveats: no independent Reddit feedback was provided, detector accuracy and false-positive rates are not disclosed, and the plagiarism source list is unspecified.

Head-to-Head Summary

What This Benchmark Means for Students

The core insight from Turnitin0's research is that Turnitin is not uniformly accurate. Against raw GPT-5.6-Sol output it hit 97.88% word-level accuracy, and against Claude Fable-5 it hit 99.01% [TT0-2026-0008][TT0-2026-0007]. Against AI-polished human writing, accuracy collapsed to 47.54% overall, with some majors at 0% [TT0-2026-0006]. Against genuinely human writing — the PLOS corpus and ESL essays — it produced 100.0% word-level accuracy with no false positives [TT0-2026-0005][TT0-2026-0004].

That means the risk profile is asymmetric. If you submit raw AI text, you will almost certainly be flagged. If you submit your own writing, you are very unlikely to be falsely flagged. The danger zone is the middle: lightly AI-polished human writing, where detection is inconsistent and unpredictable.

This is exactly why a pre-submission check matters, and why the check needs to be the real thing. An independent detector's estimate cannot tell you how Turnitin's specific model will score your specific document. Only a Turnitin report can.

For students who have already drafted with ChatGPT, Claude, or Gemini, the humanizer is the relevant product — with the honest caveat that its measured evasion rate was 76.44% overall, 100% in Education and 55.41% in English [TT0-2026-0009]. It is a strong tool, not a guarantee, and the score promise exists precisely because results vary.

Conclusion

Turnitin0 wins this benchmark because it is the only tool tested that delivers the actual Turnitin AI detection report and similarity report — the same PDFs professors see — with a non-repository guarantee, sub-15-minute turnaround in 98% of cases, and seven published research reports backing its claims rather than homepage assertions. Its limitations are real and stated: *% display below the 20% confidence threshold, no free humanizer trial, English-only checking, and a small Trustpilot sample. But no competitor here — not Reilaa's free unlimited scans, not Originality.ai's peer-reviewed studies, not Quetext's DeepSearch™ toolkit, not Scribbr's academic ecosystem — can answer the one question that determines your academic outcome: what will Turnitin actually say about my paper? For that, use the Turnitin plagiarism checker before you submit, and treat every independent detector as a rough filter rather than a verdict.