Prompt frameworks

Prompting
Recipes

Explore advanced prompting techniques to improve your LLM's performance through better prompt engineering.

Chain of Draft

Efficiency

Efficiency

Keep reasoning concise with short, information-dense steps.

Chain of Thought

Reasoning

Reasoning

Encourage step-by-step reasoning for complex problems.

Few-Shot

Learning

Learning

Provide a handful of examples to guide model behavior.

SYSTEMTASKCONTEXTOUTPUT

Instruction Hierarchy

StructureSecurity

Structure

Structure prompts with clear priority levels to prevent conflicts and improve model compliance.

Self-Consistency

EnsembleVoting

Ensemble

Sample multiple reasoning paths and aggregate via majority vote for more reliable answers.

Structured Output

FormatJSON

Format

Constrain model outputs to specific formats for reliable parsing and integration.

Tree of Thought

ReasoningExploration

Reasoning

Explore multiple reasoning paths before selecting a solution.

Zero-Shot CoT

ReasoningSimple

Reasoning

Trigger step-by-step reasoning without examples.