To try these samples out, follow these steps:
- To use the Gemini API, you'll need an API key. If you don't already have one, create a key in Google AI Studio: https://aistudio.google.com/app/apikey.
- Export a
$GEMINI_API_KEYenvironment variable with an API key with access to the Gemini generative models, or run the below commands with an environment containing this variable. - Run any sample from the
bindirectory (e.g.,dart bin/simple_text.dart).
| File | Description |
|---|---|
| chat.dart | Multi-turn chat conversations |
| code_execution.dart | Executing code |
| controlled_generation.dart | Generating content with output constraints (e.g. JSON mode) |
| count_tokens.dart | Counting input and output tokens |
| function_calling.dart | Using function calling |
| safety_settings.dart | Setting and using safety controls |
| system_instruction.dart | Setting system instructions |
| text_generation.dart | Generating text |