openai/openai-node

OpenAI Node.js / TypeScript SDK

What it is

• An official client library that lets JavaScript or TypeScript code call the OpenAI REST API.
• Generated from OpenAI’s OpenAPI spec, so the SDK mirrors the live API surface.

Core capabilities

• Text generation via the new Responses API and the classic Chat Completions API.
• Vision – send images together with text prompts and receive descriptive output.
• Streaming – receive partial results as Server‑Sent Events, useful for real‑time UI.
• File handling – helpers (toFile, toStreamingFile) make uploading fine‑tune data easy.
• Workload‑identity authentication – use short‑lived tokens from Kubernetes, Azure, GCP, or X.509 client certificates instead of static API keys.
• Realtime API – low‑latency, multimodal (text + audio) conversations over WebSocket, with function calling support.
• Azure OpenAI & Amazon Bedrock adapters – drop‑in support for the respective cloud‑provider endpoints.
• Automatic pagination, retries, timeouts, and cancellation – sane defaults that can be overridden.
• Webhook verification – built‑in methods to validate and parse OpenAI‑sent webhooks.

Typical use cases

• Building chat assistants, code generators, or any LLM‑driven feature in a web or server app.
• Processing images with LLMs (e.g., OCR, captioning) via the Vision endpoint.
• Running fine‑tuning jobs and managing uploaded training files.
• Deploying low‑latency voice assistants using the Realtime API.
• Integrating OpenAI models into Kubernetes‑based services using workload‑identity for secret‑less auth.

Installation

npm install openai   # or use Deno: import OpenAI from 'npm:openai'

Quick start (text generation with the Responses API)

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY, // optional – falls back to env var
});

const response = await client.responses.create({
  model: 'gpt-5.5',
  instructions: 'You are a coding assistant that talks like a pirate',
  input: 'Are semicolons optional in JavaScript?',
});

console.log(response.output_text);

Where to learn more

• Full method list: api.md in the repo.
• Example code: examples/ directory (e.g., multi‑turn conversations, streaming, workload‑identity).
• Official API reference: https://platform.openai.com/docs/api-reference.
• Docs for advanced topics: docs/realtime.md, docs/azure.md, and the X.509 workload‑identity guide.

Who maintains it

• Developed and maintained by OpenAI, with regular releases on npm.

---

All details are taken directly from the repository’s README; no external assumptions are added.