Tools · AWS ML Blog ·

Improve your agent’s tool-calling accuracy with SFT and DPO on Amazon SageMaker AI

Improve your agent’s tool-calling accuracy with SFT and DPO on Amazon SageMaker AI

The post shows how to improve a small language model’s tool-calling accuracy by combining supervised fine-tuning and direct preference optimization on Amazon SageMaker AI. It also covers evaluating the base model against fine-tuned variants to measure quality and guide model selection.

Read the full story at AWS ML Blog →