LLM inference

The process of using a trained language model to generate predictions, responses, or text from new inputs.