Large Language Model
A large language model is a large-scale model pretrained on extensive text or multimodal data to make probabilistic predictions over token sequences.
Definition
A large language model is a large-scale model pretrained on extensive text or multimodal data to make probabilistic predictions over token sequences.
Technical Context
An LLM is not defined by parameter count alone; data, objective, context length, post-training, tool use, and serving architecture jointly determine system behavior.
Generating an answer is distinct from actually reading a file, querying a database, or carrying out an operation. The latter capabilities require application-provided context and permitted tools. Generated text alone is not evidence that an external action occurred.