loading
loading
The date after which a model has no built-in knowledge, because its training data stops there.
A model only 'knows' what was in its pretraining and post-training data, so anything that happened after the cutoff — new library versions, recent news, a framework released last month — is invisible to it unless you feed it in. This is why agents get current info via web search, RAG, or pasted docs rather than trusting recall. Don't confuse cutoff with the current date: the model can know today's date from context while still having a cutoff months earlier, and it will confidently hallucinate about events in that gap.
Plainly
Think of Knowledge Cutoff as the brain part that guesses or decides. The date after which a model has no built-in knowledge, because its training data stops there.
In practice
Use it when model choice, prompts, latency, cost, or quality affect the product result. In practice, define the owner, input, output, and failure mode before you rely on it.