PaLM API’s Chat Service in Python – Conversation Basics

Getting Started

The Google PaLM API enables developers to create customizable chatbots and language applications using powerful language models. This AI tutorial will show you how to use the Google PaLM API’s chat service in Python.

The first step is to import the Google Generative AI module. This module contains the Python client library for the PaLM API. Next, you need to configure the API client with your API key. You can obtain the latter on the MakerSuite platform.

import google.generativeai as palm
import os

# Using an environment variable to access the saved API key
palm.configure(api_key=os.environ['PALM_API_KEY'])

Creating a New Conversation

Once you have configured the API client, create a new conversation. To do this, call the chat method on the palm module. The chat method takes a single argument, which is the message that you want to send to the API. In this case, we are sending the message “Hi!”.

# Create a new conversation
response = palm.chat(messages="Hi!")

Printing the Last Response

The chat method returns a response object. This object contains the response from the API. You can access the last attribute on the response object to print the last response generated by PaLM’s chat service.

# Print the chat service's last response
print(response.last)

Sending a Reply

You can also send a reply to the API by calling the reply method on the response object. The reply method takes a single argument, which is the prompt you want to send to the API. In this case, we are asking Google PaLM’s chat service to list three foods high in protein.

# Send a reply. Notice that the prompt is very specific to avoid inaccurate results.
response = response.reply("List only the names of 3 foods that are high in protein. No explanation.")

# Print the last response from the chat
print(response.last)

Output

The output of the chat code above is as follows:

Hi! How can I help you today?
Here are 3 foods that are high in protein:
* Chicken
* Fish
* Beans

In this short AI tutorial, we covered the following steps: Firstly, we initiated the chat service with a warm greeting. Next, we requested a list of 3 protein-rich foods. The API promptly responded with the requested list, providing only the food names without explanations, as specified. Now, with the know-how of using the Google PaLM API’s chat service in Python, you are all set to create robust chatbots and language applications.