@vertana/facade

[!CAUTION] Vertana is currently in early development for proof of concept purposes, and is not yet ready for production use. The API is subject to change, and there may be bugs or missing features.

High-level facade for Vertana providing a simple translate() function with sensible defaults. This is the recommended starting point for most applications.

Installation

deno add jsr:@vertana/facade
npm  add     @vertana/facade
pnpm add     @vertana/facade

Quick example

import { translate } from "@vertana/facade";
import { openai } from "@ai-sdk/openai";

const result = await translate(
  openai("gpt-4o"),
  "ko",
  "Hello, world!"
);

console.log(result.text);

Features

• Single function API: translate() handles the entire pipeline
• Progress reporting: Track chunking, translation, refinement stages
• Flexible options: Tone, domain, glossary, context sources
• Quality features: Refinement and best-of-N selection
• Cancellation: AbortSignal support

Options

The translate() function accepts an optional TranslateOptions object:

• mediaType: Input format ("text/plain", "text/markdown", "text/html")
• tone: Style preference (formal, informal, technical, etc.)
• domain: Subject area for terminology
• glossary: Pre-defined terminology mappings
• contextSources: External context providers
• refinement: Enable iterative quality improvement
• bestOfN: Multi-model selection for best translation
• onProgress: Callback for progress updates
• signal: AbortSignal for cancellation

For more resources, see the docs.