Text Generation
Generate text with callModel using various input formats and model configurations. Supports multiple consumption patterns including text, streaming, and structured output.
Basic Usage
The simplest way to generate text:
Input Formats
callModel accepts several input formats to match your use case.
String Input
The simplest format - a single string becomes a user message:
Message Array
For multi-turn conversations, pass an array of messages:
Multimodal
For rich content including images:
System Instructions
Set the model’s behavior with the instructions parameter:
Model Selection
Single Model
Specify a model by its OpenRouter ID:
Model Fallback
Provide multiple models for automatic fallback:
The SDK will try each model in order until one succeeds.
Response Methods
getText()
Returns just the text content after tool execution completes:
getResponse()
Returns the full response object including usage data:
Generation Parameters
Control the generation behavior:
Response Format
Request structured output:
Error Handling
Handle common error cases:
Concurrent Requests
Each callModel invocation is independent:
Next Steps
- Streaming - Stream responses in real-time
- Tools - Add tool capabilities to your generation
- Message Formats - Convert from OpenAI/Claude formats
