🚀 Integrating Gemini into Your Flutter Application

Create gemini robot 💬 chat bot 🤖 with Flutter. Show some ❤️ and star ⭐ the repo if you liked it.

contact for work, mail: chunhthanhde.dev@gmail.com

📚 Overview

Integrate Google's Gemini AI into your Flutter app with ease using the flutter_gemini package. This guide will help you enhance your app's functionality by leveraging Gemini's advanced AI features, including multilingual support, image and audio processing, and Natural Language Processing (NLP).

🌟 Key Features of Gemini

• Stream Generate Content: Handle partial results with streaming without waiting for the complete response.
• Text-only Input: Complete and summarize texts using NLP.
• Text-and-Image Input: Use the gemini-pro-vision model for tasks involving both text and images.
• Multi-Turn Conversations (Chat): Engage in dynamic conversations with follow-up questions.
• Token Counting: Convert information into floating-point numbers.
• Model Info: Retrieve details about the model, including version and input token limits.
• EmbedContents and BatchEmbedContents: Transform information into numerical arrays for learning.

🏁 Implementation Steps

1. Obtain an API Key 🔑 Create an account at Google Gemini.

   • Click on "Get API Key,"

     

   • select a Google Cloud project,

     

   • and generate your API key.

     

   • Save this key for later use.

     

2. Add Dependency 📦

   

   • Add the flutter_gemini package to your project by running:

     flutter pub add flutter_gemini

3. Initialize Gemini ⚙️

   • In your main.dart file, initialize the Gemini package with your API key:

     void main() {
       Gemini.init(apiKey: 'YOUR_API_KEY_HERE');
       runApp(const MyApp());
     }

4. Create an Instance and Make Requests 🤖

   • Use the following code to create an instance of flutter_gemini and handle text or image input:

     final _gemini = Gemini.instance;
     _gemini.textAndImage(
       text: 'Your query here',
       images: [imageFile.readAsBytesSync()],
     );

5. Handle Responses 🔍

   • Manage responses from Gemini with attributes such as FinishReason and SafetyRating:

     if (value?.finishReason != 'STOP') {
       setResponse(value?.output ?? "Invalid User Query");
     }
     if (value?.SafetyRating != 'HarmCategory') {
       setResponse(value?.candidates[0].safety_ratings);
     }

🚀 Demo

You can see the response from the gemini here: https://www.youtube.com/watch?v=SUzw8YPLYZM

🎉 Benefits

• Easy Integration: Simplifies adding AI features with clear documentation.
• Cross-Platform Compatibility: Supports Android, iOS, web, macOS, and Windows from a single codebase.
• Reduced Time to Market: Speeds up the integration of AI features into your app.

🌟 Conclusion

Integrating Google Gemini into your Flutter app with the flutter_gemini package opens up new possibilities for enhancing user experience. Utilize Gemini's powerful AI capabilities to add multilingual support, process images and audio, and implement advanced chatbot functionality.

📚 References

• flutter_gemini Package
• GitHub Repository
• Google AI

Feel free to like and share your thoughts in the comments if you found this guide helpful! 👍