@charivo/llm-client-openai

Direct OpenAI browser LLM client for Charivo.

Use this only for local development and testing. It enables browser-side access by passing dangerouslyAllowBrowser: true to the underlying provider.

For production, prefer @charivo/llm-client-remote.

Install

pnpm add @charivo/llm-client-openai

Usage

import { createOpenAILLMClient } from "@charivo/llm-client-openai";

const client = createOpenAILLMClient({
  apiKey: process.env.NEXT_PUBLIC_OPENAI_API_KEY!,
  model: "gpt-4.1-nano",
});

const message = await client.call([
  { role: "user", content: "Hello" },
]);

Config

The client reuses OpenAILLMConfig from @charivo/llm-provider-openai:

• apiKey
• model?
• temperature?
• maxTokens?