Back
PangramTRY IT

Pangram

Pangram is an AI detector for text that also checks images and video. It checks writing from ChatGPT, Claude, Gemini and other models, marks which passages look human-written and which look AI-assisted or AI-generated, and runs a similarity check for plagiarism on paid plans. Independent studies from researchers at the University of Chicago and the University of Maryland ranked it the most accurate AI text detector they tested.

Teachers, editors, publishers, recruiters and trust and safety teams use it when they need to review submitted work at volume.

Text detection

Paste text or upload a document, and Pangram returns an overall result plus passage-level highlights. That view helps when a document mixes original writing with AI edits. It works in more than 20 languages and reads scanned documents with OCR. Pangram trains it on the latest models, including GPT, Claude, Gemini, Grok and Llama. Pangram reports a false positive rate of about 1 in 10,000 on human writing.

Images and video

Pangram Image 1.0 checks common image files for signs of AI generation, and a newer video detector checks uploaded clips.

Where you can use it

Integrations bring checks into Google Docs and Gmail. Paid plans can also scan your social media and Substack feeds automatically. Developers can call the API or connect through MCP, with per-use rates for each type of check and a discount for bulk text.

Pricing

Pangram is free to try with 20 credits a day. The Individual plan starts with a 7-day free trial and adds monthly scan allowances, similarity detection, file upload and the integrations. The Professional plan raises the limits and includes API credits. Organizational plans cover schools and companies.

Can Pangram prove that someone used AI?

No detector can prove authorship on its own. Use a Pangram result as a reason to look closer at a document and its history, especially in schools or hiring, where a false accusation is costly.

Alternatives to Pangram

  • AI Aware: Detects AI-generated text, images, audio and deepfake video in one place, with free checks and credit packs.
  • Quetext: Checks writing for AI patterns and plagiarism, then helps rewrite, cite and review text through web, API and MCP tools.
  • TruthScan: Screens images and PDFs for signs of AI generation or tampering, with confidence scores and explanations.
  • DetectVideo AI: Checks whether a video is likely AI-generated from an upload or a supported public link.