Early Beta — The Web SDK is in early beta. APIs may change between releases.
Overview
The TextGeneration.generate() method with default options provides the simplest way to generate text. For a one-line approach, just pass a prompt and get a result.
Basic Usage
import { TextGeneration } from '@runanywhere/web'
const result = await TextGeneration.generate('What is the capital of France?')
console.log(result.text) // "Paris is the capital of France."
API Reference
await TextGeneration.generate(
prompt: string,
options?: LLMGenerationOptions
): Promise<LLMGenerationResult>
Parameters
| Parameter | Type | Description |
|---|
prompt | string | The user’s message or question |
options | LLMGenerationOptions | Optional generation settings |
Returns
| Type | Description |
|---|
Promise<LLMGenerationResult> | Result with text and performance metrics |
Throws
| Error Code | Description |
|---|
NotInitialized | SDK not initialized |
ModelNotLoaded | No LLM model loaded |
GenerationFailed | Generation failed |
Examples
Simple Q&A
const capital = await TextGeneration.generate('What is the capital of Japan?')
// capital.text: "Tokyo is the capital of Japan."
const math = await TextGeneration.generate('Calculate 15% of 200')
// math.text: "15% of 200 is 30."
With Options
const result = await TextGeneration.generate('Write a haiku about coding', {
maxTokens: 50,
temperature: 0.8,
})
console.log(result.text)
console.log(`${result.tokensPerSecond.toFixed(1)} tok/s | ${result.latencyMs}ms`)
Error Handling
import { TextGeneration, SDKError, SDKErrorCode } from '@runanywhere/web'
try {
const result = await TextGeneration.generate('Hello')
} catch (err) {
if (err instanceof SDKError) {
switch (err.code) {
case SDKErrorCode.NotInitialized:
console.error('Initialize the SDK first')
break
case SDKErrorCode.ModelNotLoaded:
console.error('Load a model first')
break
default:
console.error(`SDK error [${err.code}]: ${err.message}`)
}
}
}
Simple vs Full Generation
| Feature | Simple (defaults) | Full (generate() with options) |
|---|
| Return type | LLMGenerationResult | LLMGenerationResult |
| Metrics | Yes | Yes |
| Options | Defaults | Customizable |
| System prompt | None | Yes |
| Use case | Quick prototyping | Production apps |
For quick prototyping, use generate() with just a prompt. Switch to adding options when you need
custom temperature, system prompts, or token limits.
