What is a prompt?

A prompt is input you send to a model in order for it to generate a response, in a process known as inference. For example, you could send the following prompt to a model.

What is Avebury stone circle?

When you run the prompt in the playground, the playground shows a response from the model that is similar to the following.

Avebury stone circle is a Neolithic monument located in Wiltshire, England. 
It consists of a massive circular bank and ditch, with a large outer circle of standing stones
that originally numbered around 100.

The Explore mode playground provides quick start prompts that illustrate the kinds of prompt that you can send to a model.

The Explore playground also lets you try multimodal prompts, which are prompts that include images and text. This allows you to pass an image to a model and ask questions such as What's in this image?. Not all models support multimodal prompts.

The actual response that you get for a prompt depends on the model you use.

With Amazon Bedrock models you can use inference parameters to influence the response from a model. For example, you can use the temperature inference parameter to filter out lower probability responses. To experiment with inference parameters, or use an Amazon Bedrock component, such as a guardrail, you need to create a Bedrock Studio app. You can create a chat app or a Prompt Flows app. If you create a Prompt Flows app, you can also create prompts that you can customize for different use cases. For more information, see Use prompts with your Flows app.

Prompt engineering guides

Amazon Bedrock Studio provides models from a variety of model providers. Each provider provides guidance on how to best create prompt for their models.

For general guidelines about creating prompts with Amazon Bedrock, see General guidelines for Amazon Bedrock LLM users.