OpenAI API - first call, streaming, and cost control

1. Get the key

Create an account at platform.openai.com, go to API Keys -> Create new secret key. Add credit in Billing (minimum US$5). Store the key in an environment variable, never in the frontend.

2. First call (TypeScript)

import OpenAI from "openai";
const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });
const res = await client.chat.completions.create({
  model: "gpt-5-mini",
  messages: [{ role: "user", content: "Explain RLHF in 2 sentences." }],
});
console.log(res.choices[0].message.content);

3. Streaming

const stream = await client.chat.completions.create({
  model: "gpt-5-mini",
  messages: [{ role: "user", content: "Write a haiku about TypeScript" }],
  stream: true,
});
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}

4. Structured outputs

const res = await client.chat.completions.create({
  model: "gpt-5",
  messages: [{ role: "user", content: "Extract name and email: Jane <jane@example.com>" }],
  response_format: { type: "json_schema", json_schema: {
    name: "Contact",
    schema: { type: "object", required: ["name","email"],
      properties: { name: {type:"string"}, email: {type:"string"} } },
    strict: true,
  }},
});

5. Cost control

• •Use `gpt-5-mini` as the default; move to `gpt-5` only where reasoning quality matters
• •Enable prompt caching for 50-90% discounts on repeated prefixes
• •Truncate old history (keep the last N messages)
• •Rate-limit on your side so the monthly budget cannot spike