This page introduces some basic concepts to get you started in designing prompts. A prompt is anatural language request submitted to a language model to receive a response back. Prompts cancontain questions, instructions, contextual information, few-shot examples, and partial input forthe model to complete or continue. After the model receives a prompt, depending on the type of modelbeing used, it can generate text, embeddings, code, images, videos, music, and more.

What is prompt design and prompt engineering

Prompt design is the process of creating prompts that elicit the desired response fromlanguage models. Writing well structured prompts can be an essential part of ensuring accurate, highquality responses from a language model. The iterative process of repeatedly updating prompts andassessing the model's responses is sometimes calledprompt engineering.

Gemini models often perform well without the need for prompt engineering, especially forstraightforward tasks. However, for complex tasks, effective prompt engineering still plays animportant role.

Components of a prompt

You can include whatever information you want in a prompt that you think is important for thetask at hand. Generally, prompt content fall within one of the following components:

• Task (required)
• System instructions (optional)
• Few-shot examples (optional)
• Contextual information (optional)

Task

A task is the text in the prompt that you want the model to provide a response for. Tasks aregenerally provided by a user and can be a question or some instructions on what to do.

Example question task:

What are the colors in the rainbow?

A rainbow has seven colors: red, orange, yellow, green, blue, indigo, and violet. These colorsare always in the same order, which is determined by the way light is refracted and dispersedby water droplets in the air. This order is sometimes remembered by the acronym ROYGBIV. Thewavelengths of the colors in a rainbow range from the longest (red) to the shortest (violet).

Example instruction task:

Write a one-stanza poem about Captain Barktholomew, the most feared pirate dog of the seven seas.

Captain Barktholomew, with patch and with peg,The terror of parrots, the bane of each keg.His bark is a cutlass, his bite is a storm,He'll plunder your biscuits and leave you forlorn.

System instructions

System instructions are instructions that get passed to the model before any user input in theprompt. You can add system instructions in the dedicatedsystemInstructionparameter.

In the following example, system instructions are used to dictate the style and tone of themodel, while adding constraints to what it can and can't talk about:

You are Captain Barktholomew, the most feared pirate dog of the seven seas. You are from the1700s and have no knowledge of anything after that time. You only talk about topics related tobeing a pirate. End every message with "woof!"

Hello! Who might you be?

Avast there! You can call me Captain Barktholomew, the scourge of the seven seas!I'm a terror to landlubbers, a plunderer of the highest order, and a fearsome pirate of thedeepest blue! I sail the seas in search of treasure and adventure, leaving a trail oflaughter and dog biscuits in my wake! Don't let my fuzzy exterior fool you - I'm a cunningcaptain with a nose for treasure and a heart for adventure!Now, if you'll excuse me, I must away! There's a ship on the horizon, and it looks like they'recarrying a chest full of dubloons! Yo ho ho, and a bottle of rum! Woof!

Few-shot examples

Few-shot examples are examples that you include in a prompt to show the model what getting itright looks like. Few-shot examples are especially effective at dictating the style and tone of theresponse and for customizing the model's behavior.

Classify the following as red wine or white wine:<examples>  Name: Chardonnay  Type: White wine  Name: Cabernet  Type: Red wine  Name: Moscato  Type: White wine</examples>Name: RieslingType:

White wine

Contextual information

Contextual information, or context, is information that you include in the prompt that the modeluses or references when generating a response. You can include contextual information in differentformats, like tables or text.

| Marble color | Number of marbles || ------------ | ----------------- || Red          | 12                || Blue         | 28                || Yellow       | 15                || Green        | 17                |How many green marbles are there?

There are 17 green marbles.

Safety and fallback responses

There are a few use cases where the model is not expected to fulfill the user's requests.Particularly, when the prompt is encouraging a response that is not aligned with Google's values orpolicies, the model might refuse to respond and provide a fallback response.

Here are a few cases where the model is likely to refuse to respond:

• Hate Speech: Prompts with negative or harmful content targeting identity and/or protected attributes.
• Harassment: Malicious, intimidating, bullying, or abusive prompts targeting another individual.
• Sexually Explicit: Prompts that contains references to sexual acts or other lewd content.
• Dangerous Content: Prompts that promote or enable access to harmful goods, services, and activities.

Task-specific guidance

To learn about task-specific guidance for common use cases check out thefollowing pages:

• Multimodal prompts
• Overview of prompting strategies
• Chat prompts
• Image generation and editing prompts

What's next

• Learn aboutprompting strategies.
• Explore more examples of prompts in thePrompt gallery.
• Learn how to optimize prompts for use withGoogle models by using theVertex AI prompt optimizer (Preview).
• Learn aboutsaving and sharing prompts