@llm-optimize/openai

OpenAI adapter for @llm-optimize/core. Works with OpenAI, Azure OpenAI, Groq, Together AI, and any OpenAI-compatible API.

Install

npm install @llm-optimize/openai @llm-optimize/core

Usage — OpenAI

import { LLMOptimizer } from "@llm-optimize/core";
import { createOpenAIProviderClient } from "@llm-optimize/openai";

const ai = new LLMOptimizer({
  provider: "openai",
  providerClientFactory: {
    create: () => createOpenAIProviderClient({
      apiKey: process.env.OPENAI_API_KEY,
    }),
  },
});

const response = await ai.chat({
  model: "gpt-4o",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "Write a short poem about TypeScript." },
  ],
  temperature: 0.7,
  maxTokens: 512,
});

console.log(response.content);
console.log(response.usage); // { inputTokens, outputTokens, totalTokens }

Usage — Azure OpenAI

import { createOpenAIProviderClient } from "@llm-optimize/openai";

const client = createOpenAIProviderClient({
  apiKey: process.env.AZURE_OPENAI_API_KEY,
  baseUrl: "https://<your-resource>.openai.azure.com/openai/deployments/<deployment>",
});

Usage — Groq / Together AI (OpenAI-compatible)

const client = createOpenAIProviderClient({
  apiKey: process.env.GROQ_API_KEY,
  baseUrl: "https://api.groq.com/openai/v1",
});

Supported Models

| Model | ID | |---|---| | GPT-4o | gpt-4o | | GPT-4o mini | gpt-4o-mini | | GPT-4 Turbo | gpt-4-turbo | | GPT-3.5 Turbo | gpt-3.5-turbo | | o1 | o1 | | o1-mini | o1-mini |

What's Included

• createOpenAIProviderClient(options) — creates a fetch-based OpenAI client
• OpenAIProvider — class wrapper around any OpenAIClient
• OpenAIClient interface — implement this to bring your own HTTP client
• OpenAIProviderOptions — { apiKey?, baseUrl? }