How can I parse a streamed JSON array response using Node.js OpenAI API v4.0.1?

Below is an example using the built-in async iteration support in the OpenAI Node API. With v4.0.1 the client returns an async iterable that already parses the JSON for you. Instead of using an external library like stream-json, you can consume the stream directly and process each JSON element (chunk) as it’s received:

JAVASCRIPTCopied
import OpenAI from "openai";

const openai = new OpenAI();

async function main() {
  const stream = await openai.chat.completions.create({
    model: "gpt-4o-mini",
    messages: [{ role: "user", content: "Say this is a test" }],
    stream: true,
  });

  // Process each streamed chunk as a JSON element
  for await (const chunk of stream) {
    // Each chunk is already parsed as a JSON object
    const text = chunk.choices[0]?.delta?.content;
    if (text) {
      console.log(text);
    }
  }
}

main();

Key Points

• Built-In Parsing: The Node API client automatically handles the JSON parsing of the server-sent events, so you don’t need to manually transform the stream into JSON objects.
• Async Iteration: The use of a for await...of loop allows you to process individual elements (chunks) as they arrive.
• Type Compatibility: Because the client returns properly typed objects, you avoid issues with external stream parsers like stream-json.

This approach should meet your requirements of parsing a streamed chat completion response and processing individual JSON elements without running into type mismatches.