Sample prompts

Explore these sample prompts to understand how Large Language Models (LLMs) can be utilized in your service flow.

To get the most out of these examples, you can experiment with different LLMs and adjust the values in the Parameters section. This will allow you to see how the generated output changes based on your inputs.

Table 1. Sample prompts
Example Context Input
Generate loan acceptance letter Generate an acceptance letter to the loan request provided as input.

Customer:{{customerName}}

Reference:{{ref}}

Amount:{{amount}}
Generate translation The input may be in another language, translate it into English {{message}}
Generate sentiment Analyze the sentiment of a customer review provided as input and give it a note between 0 and 5 included, where 0 means "very unsatisfied" and 5 means "very satisfied". The output must only be made of a single digit. {{customerReview}}

Sample prompts for Llama 3 models

The watsonx.ai™-supported Llama 3, 3.1, 3.2, and 3.3 models are optimized for dialog use cases and fine tuned for a specific prompt format, which is described in the Prompt text section in Sample: Converse with Llama 3 External link opens a new window or tab. The generative AI task uses this specific prompt format, which means that you do not need to create the prompt text with the specific prompt template.

When you use a Llama-3-3-70b-instruct model, the Gen AI page looks like the following example:
The Gen AI page for a llama-3-3-70b-instruct model shows the following options: Write a sentence about this topic. in the Prompt -> Context area, {{topic}} in the Prompt -> Input area, and winter tires as a value in the Variables -> Value column.

When you examine the RAW prompt, notice that the specific prompt template is used.

RAW prompt for a llama-3-3-70b-instruct model shows that the specific prompt template is used.

Sample prompts for non-Llama models

Non-Llama models use a generic template with an Input: stop sequence.

For a non-Llama model, the previous example shows the following options in the Gen AI page:

Non-Llama models use a generic template with an Input: stop sequence.

The corresponding RAW prompt looks like this:

RAW prompt for a non-llama model that uses a generic template with an Input: stop sequence.

Sample prompts for curated models

Alternatively, for a model that uses a generic template but requires a specific prompt template, you can define the template along with the prompt in the Prompt -> Context area. You do not need to enter any values in the Prompt -> Input area.

For example, in a curated model scenario, you can use the prompt template that is required in the model. In the following screen capture, the Prompt -> Input area is empty, and a prompt template is provided in the Prompt -> Context area.

The Gen AI page for a curated model that uses a generic template shows the following options: an empty Prompt -> Input area and a prompt template in the Prompt -> Context area.

The corresponding RAW prompt looks like this:

RAW prompt for a curated model that uses a generic template.
Important: The text generation request in the generative AI tasks has an Input: stop sequence.