sft

What problem does it solve?

SFT helps tailor pre-trained LLMs to follow instructions by training on high-quality instruction-response data, reducing misalignment and improving task performance.

Core Features & Use Cases

• Unsloth-optimized training for faster, memory-efficient fine-tuning.
• Thinking model patterns that encourage internal reasoning demonstrations during evaluation and generation.
• Dataset and template tooling formats for instruction-response and chat formats, plus prompt construction to streamline fine-tuning workflows.

Quick Start

Fine-tune an instruction-following model by configuring SFTTrainer with your dataset and a LoRA setup, then run training.