2026-10-06 07:51:44
We have updated two of our most popular tools: the plagiarism checker and the AI detector.
Plagiarism checker
- Faster results. A check usually takes 7–15 seconds instead of about 45, and only a couple of seconds when similar pages have been checked before.
- Finds rewritten text. If someone retells another article in their own words, the checker still finds the source and shows the matches. Such text no longer gets a high originality score.
- More sources. Results list up to three times more sources, so the original is missed less often.
- Fewer false matches. Your own text no longer loses points because of common phrases.
- Precise highlighting. Matching fragments are marked exactly, and you can jump to every source in the list.
- PDF report. It contains the highlighted matches, the list of sources, the check date and a link to the result: easy to attach to your work or send to a client.
- What next. After the check, one click takes your text to the paraphrasing tool, the Humanizer or the AI detector.
AI detector
- Recognises modern AI models: ChatGPT, Claude, Gemini, DeepSeek, GigaChat, Grok, Qwen, Kimi and GLM.
- Fewer false accusations. 99% of texts written by people are recognised as human: essays, articles, news, reviews and posts. Classic literature, the Bible and constitutions are not mistaken for AI either.
- 15 languages: English, Spanish, Portuguese, German, French, Italian, Russian, Ukrainian, Polish, Turkish, Bulgarian, Chinese, Arabic, Urdu and Indonesian. Texts in other languages are checked too.
- Results in seconds. Suspicious fragments are highlighted, and the report can be downloaded as a PDF.
How accurate is it?
We tested the detector on thousands of texts it had never seen during training: texts written by people before ChatGPT appeared, and texts generated by AI models.
| AI model | Detected as AI |
|---|---|
| ChatGPT (GPT-6 Sol, GPT-6 Luna, GPT-5.6 and earlier versions) | 97.7–100% |
| Claude (Opus 5, Sonnet 5.5) | 100% |
| Gemini (3.8 Flash, 3.5 Flash Lite, 2.0) | 99.5–100% |
| DeepSeek (V4 Pro, V4.1 Flash) | 100% |
| GigaChat | 98.3% |
| Grok 4.7, Qwen 3.8, Kimi K3, GLM 5.3 | 100% |
| All AI models | 98.8% |
| Texts by people | Recognised as human |
|---|---|
| English | 98.8% |
| Spanish | 98.4% |
| Portuguese | 97.7% |
| German | 98.5% |
| French | 98.5% |
| Italian | 98.0% |
| Russian | 100% |
| All texts by people | 99% |
New AI models appear all the time. Texts from models the detector has not met yet are recognised in about 95% of cases, and we add new models regularly. The detector shows a probability, not a verdict: assess a disputed text together with other signs.