Support agents.One key, many models

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://tokens.flex.ai/v1",
  apiKey: process.env.FLEXAI_API_KEY,
});

const audio = await fetch("/call.mp3").then((r) => r.blob());
// search() → your vector store lookup (pgvector, Pinecone, …)

// 1. Transcribe: same key, swap the model anytime
const transcribe = await client.audio.transcriptions.create({
  model: "whisper-large-v3-turbo",
  file: audio,
});

// 2. Read attachments: same key, swap the model anytime
const readAttachments = await client.chat.completions.create({
  model: "PaddleOCR-VL",
  messages: [{ role: "user", content: transcribe.text }],
});

// 3. Retrieve: same key, swap the model anytime
const retrieve = await client.embeddings.create({
  model: "bge-m3",
  input: readAttachments.choices[0].message.content,
});

// 4. Respond: same key, swap the model anytime
const respond = await client.chat.completions.create({
  model: "gpt-oss-20b",
  messages: [{ role: "user", content: await search(retrieve.data[0].embedding) }],
});

// 5. Speak: same key, swap the model anytime
const speak = await client.audio.speech.create({
  model: "Kokoro-82M",
  voice: "alloy", input: respond.choices[0].message.content,
});