openai

OpenAI's toolkit for building Model Context Protocol integrations.

GPT-powered triage tool for processing BugCrowd bug bounty submissions.

Next.js photobooth app built on the OpenAI Image API — a reference for image-generation product flows.

Developer toolkit to migrate applications from the legacy OpenAI Completions/Chat Completions APIs to the unified Responses API, guided by Codex CLI.

Example MCP server compatible with OpenAI Deep Research. Reference implementation for connecting custom data sources to Deep Research.

Lightweight Android capture developer tool for macOS. Capture screenshots and screen recordings from Android devices directly on your Mac.

Open-source starter kit from OpenAI for building knowledge retrieval (RAG) systems on top of the OpenAI API.

TypeScript and JavaScript SDK for OpenAI Guardrails — runtime safety controls for LLM applications.

Official OpenAI Python library for adding input/output guardrails to LLM apps — block jailbreaks, PII leaks, and off-topic requests.

Example AI meeting assistant using the OpenAI Realtime API — listens to a call and updates a Kanban board live.

Reference hardware project for a physical AI assistant — illuminates a mirror with ambient info and runs voice interactions.

Customer-support agent reference built with Next.js and the OpenAI Responses API plus File Search — clone-and-go template.

Official Python SDK for OpenAI's ChatKit, a toolkit for building chat-style AI apps.

Browser viewer for inspecting harmony chat data and Codex agent sessions.

Official OpenAI toolkit for social scientists to measure quantitative attributes in text, images, or audio via GPT.

OpenAI's collection of security automation bots: triage, analysis, and response workflows powered by LLMs.

Official Ruby SDK for the OpenAI API. Chat, embeddings, audio, and assistants from idiomatic Ruby.

Demo that uses the OpenAI Realtime API and tool calling to navigate a 3D solar system by voice.

OpenAI's reference implementation of sparse autoencoders for interpretability research on large language models.

Official command-line interface for the OpenAI API, written in Go.

Demo of a UI testing agent built on OpenAI's Computer Use model and the Responses API. Drives a browser to test apps end-to-end.

Reference app for the OpenAI Responses API. Shows streaming, tool calls, and conversation state in a small TypeScript codebase.

React component for the OpenAI Realtime API. Drop in voice-to-voice chat with the right audio handling, interruption, and event wiring.

Reference apps for OpenAI's Structured Outputs feature. Patterns for getting reliable JSON out of GPT models via schemas.

React components and primitives for building UIs inside OpenAI's Apps SDK — the platform for shipping interactive apps inside ChatGPT.

Base Docker image used by OpenAI Codex environments — preinstalled toolchains for many languages.

Reference Node.js/JavaScript client for the OpenAI Realtime API — voice agents and live conversations.

Official GitHub Action that runs OpenAI Codex in your CI to review, fix, or implement code automatically.

Node.js example app from OpenAI's quickstart tutorial — minimal, runnable reference for building your first OpenAI-powered app.

OpenAI's code evaluation dataset and harness from the Codex paper — 164 hand-written Python problems for benchmarking LLMs on code.

Lightweight, powerful framework for multi-agent workflows and voice agents in JavaScript/TypeScript — built by OpenAI for production.

Official Go library for the OpenAI API — chat, completions, embeddings, audio, batch, files, and the Responses API.

Reference demo of a customer-service workflow built with the OpenAI Agents SDK — multi-agent handoffs, tools, and a sample frontend.

React app for inspecting, building, and debugging with the OpenAI Realtime API — see WebSocket events, run voice agents, iterate fast.

OpenAI's renderer for the Harmony response format used by gpt-oss — chat templates, tool calls, and structured outputs for open-weight GPT models.

OpenAI's challenge: train the smallest language model that fits in 16MB. Best model wins — a benchmark for tiny-LM engineering.

Reference patterns for building voice agents on the OpenAI Realtime API — handoffs, tool use, and low-latency speech-to-speech.

OpenAI's generative model for 3D assets — produce 3D objects from text prompts or images, exported as meshes or implicit functions.

OpenAI's fast BPE tokenizer — count tokens, encode/decode for GPT models, written in Rust with Python bindings.

OpenAI's point-cloud diffusion model for 3D synthesis — generate 3D point clouds from text in roughly a minute on a single GPU.

Official Python library for the OpenAI API — chat, completions, embeddings, audio, batch, files, fine-tuning, and the Responses API.

OpenAI's educational multi-agent orchestration framework — ergonomic primitives for handoffs between agents with shared context.

OpenAI's robust speech recognition model trained on 680k hours of multilingual audio, with strong performance on accents and technical language.

OpenAI's vision-language model that predicts the most relevant text snippet for a given image, trained on 400M image-text pairs from the web.

OpenAI's framework for evaluating LLMs and LLM systems, with an open-source registry of benchmarks the community can extend.

Symphony transforms project management by creating isolated, autonomous implementation runs where AI coding agents handle complete tasks from Linear boards while providing proof of work including CI status, PR reviews, and complexity analysis. Teams can manage work at a high level instead of supervising individual coding agents.

A lightweight framework for building multi-agent workflows in Python. Supports agent handoffs, tools, guardrails, human-in-the-loop interactions, and works with 100+ LLMs beyond just OpenAI.

A coding agent from OpenAI that runs locally in your terminal, providing AI-powered assistance for development tasks. Works with ChatGPT plans or API keys and offers multiple deployment options including IDE integration and desktop app.

The OpenAI Model Spec — a public document defining how OpenAI's models should behave, the principles that govern them, and the rules they follow.