Chat Conversations with Gemini

Maintain context over multiple messages in a conversation using Gemini's chat functionality.

Typescript

import { GoogleGenerativeAI } from "@google/generative-ai";

const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY);
const model = genAI.getGenerativeModel({ model: "gemini-2.0-flash" });

// Start a chat session
const chat = model.startChat({
  history: [
    {
      role: "user",
      parts: "Hello, Gemini."
    },
    {
      role: "model",
      parts: "Hi there! How can I assist you today?"
    }
  ]
});

// Continue the conversation
const result1 = await chat.sendMessage("I have 3 cats. How many paws are in total?");
console.log(result1.response.text());

const result2 = await chat.sendMessage("Can you translate that answer into French?");
console.log(result2.response.text());

Python

from google import genai

# Initialize the model
model = genai.GenerativeModel("gemini-2.0-flash")

# Start a chat session with history
chat = model.start_chat(history=[
    {"role": "user", "parts": "Hello, Gemini."},
    {"role": "model", "parts": "Hi there! How can I assist you today?"}
])

# Continue the conversation
response1 = chat.send_message("I have 3 cats. How many paws are in total?")
print(response1.text)

response2 = chat.send_message("Can you translate that answer into French?")
print(response2.text)

Chat Configuration Options

• history: Array of previous messages
• context: Additional context for the conversation
• examples: Example interactions to guide the model

Best Practices for Chat

1. Keep conversation history concise
2. Provide clear context when needed
3. Use examples to guide model behavior