LLM (large language model)
A neural network trained on huge text corpora to predict the next token — the engines behind ChatGPT, Claude, Gemini, etc.
A large language model (LLM) is a neural network trained on huge corpora of text with the objective of predicting the next token in a sequence. GPT-4, Claude, Gemini, Llama, and DeepSeek are all LLMs.
LLMs are what AI detectors are detecting. The same machine-learning principles that let an LLM generate fluent text also leave statistical traces in that text — patterns of token probability, sentence rhythm, lexical preference — which is what detection algorithms key on.
AI Detector API is built to detect output from all major LLM families. As new models ship, we retrain on representative samples so detection accuracy stays current.
Related terms
- Token— The smallest unit of text a language model processes — usually a word or a piece of a word.
- Perplexity— How surprised a language model is by a given piece of text — lower means the text looks more model-generated.
- AI detection— The task of identifying text that was written by a large language model rather than a human.
Move from definition to code
Free 1,000 requests/month — no credit card. Be detecting AI text in 5 minutes.