VID 20240920 040106

This is a video about building a generative AI application for a use case supporting productivity with IBM WatsonX.AI, featuring IBM Granite.
Here I am explaining about building a generative AI application called "Text Generation AI App," which is a simple text-to-text generation application. It is just like a chatGPT or chatBot kind of application where we can give prompts to the model and get Output from the AI.
This application has three features where we can give different types of prompts according to our requirements.
1)Chat: Customize your chat
Before you start chatting, you can update the current settings and ground the chat with documents. To upload documents or an image, click next to the input field.
2) Structured: You can add an imperative statement that tells the model what to do. Including descriptive qualifiers can improve the quality of the results. For example: "Summarize the text in a list, using a concise, professional tone.
(OR)
You can show the model how to format the response by entering one or more pairs of example input and desired output. You can change the Input and Output labels to provide more structure in this section and the Try section.

Summarization
Classification
Generation
Extraction
Question answering
Code Generation and Translation
Translation (English to French & Spanish, etc)

3)Freeform

Enter your prompt text. Provide instructions and examples to show the model what to do.