An ML engineer must fine-tune a large language model (LLM) for text summarization using Amazon SageMaker and must use a low-code/no-code workflow. Which option satisfies these constraints?

The correct answer is to use SageMaker Autopilot to fine-tune an LLM that is deployed through SageMaker JumpStart. This option directly addresses the requirement for a low-code/no-code workflow for fine-tuning an LLM. SageMaker JumpStart provides pre-trained models, including LLMs, that can be deployed and fine-tuned with minimal code. SageMaker Autopilot automates the machine learning process, including hyperparameter tuning, making it a low-code solution for fine-tuning. Using SageMaker Studio to fine-tune an LLM on EC2 instances would involve more coding and manual setup, not meeting the low-code/no-code constraint. SageMaker Autopilot is not designed for deploying models via custom API endpoints or directly onto EC2 instances; its output is typically a SageMaker endpoint.