Prompt Repetition Optimization

What problem does it solve?

Prompt repetition improves LLM performance on non-reasoning tasks by enabling each prompt token to attend to every other token, reducing context-attention limitations.

Core Features & Use Cases

• Best for non-reasoning tasks such as multiple-choice questions with options-first formatting, list navigation, and simple fact retrieval.
• Variants supported: simple repetition, verbose repetition, and triple repetition for flexible accuracy gains.
• Compatibility: drop-in with existing prompting flows and no need to modify model or tooling.

Quick Start

Use a non-reasoning prompt and apply prompt repetition by sending the prompt twice (the base prompt followed by the same prompt) to the model.