Apply AI

Gemini in Flutter

Wiring Google's Gemini models into Flutter through the REST API.

Problem: Flutter teams want generative AI without adopting a heavy, fast-moving SDK.
Decision: Integrated the Gemini REST API directly in Flutter and documented the pattern for the community.
Result: A practical guide and demo for Flutter developers.

Flutter
Dart
Gemini
REST

Sometimes the right call is to skip the SDK. Talking to Gemini over its REST API keeps a Flutter app light and the dependency surface small, which matters when the model side changes every month.

The shape is a plain HTTPS POST:

final res = await http.post(
  Uri.parse(
    "https://generativelanguage.googleapis.com/v1beta/models/gemini-pro:generateContent?key=$apiKey",
  ),
  headers: {"Content-Type": "application/json"},
  body: jsonEncode({
    "contents": [
      {
        "parts": [
          {"text": prompt},
        ],
      },
    ],
  }),
);

From there it is ordinary Dart: decode the response, handle errors, and render the result into the UI. I wrote the full version up for developers who want the pattern, not a black box.

Read the guide