Codex
In plain wordsA name OpenAI has used twice: first for a 2021 AI model that wrote code and powered the original GitHub Copilot, then from 2025 for its family of coding agents.
- 1
Coding agents (2025 onward)
OpenAI's family of coding agentscoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed., introduced in 2025. Codex CLI (April 2025) is an open-source agent that runs on the developer's own machine; the cloud version (May 2025) takes many assignments at once, gives each a separate sandboxsandbox, n.: A sealed-off computer space where an AI agent can run programs and change files without being able to harm the real system, reach private data or contact the outside world unless allowed. holding a copy of the user's repository, and hands back proposed changes together with the terminal logs and test results that back them up. That cloud agent first used codex-1, an o3 reasoning modelreasoning model, n.: An AI model trained to work through a problem step by step in hidden 'thinking' text before it answers. Letting it think longer generally makes it better at math, coding and planning. further trained for programming work, and had no internet access while a task ran; later releases added editor extensions, code review and newer models. Its launch introduced the AGENTS.md convention for repository instructions (see agent instruction fileagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session.).
Kick the refactor off in Codex and review the diff after lunch.
- 2
Code model (2021–2023)
A GPT-3GPT-3, prop. n.: A huge 2020 OpenAI language model that could perform new tasks simply from instructions or a few examples written into its prompt, without retraining. It made prompting a new way of programming. descendant fine-tuned on public GitHub code, described in the 2021 paper that introduced the HumanEvalHumanEval, prop. n.: A set of 164 small programming exercises, each described in plain English and checked by hidden tests, used to see whether an AI can write working Python code. benchmark — where it solved 28.8% of problems with one sample, against 0% for GPT-3 — and used to power the first version of GitHub CopilotGitHub Copilot, prop. n.: An AI assistant from GitHub, built into programmers' code editors, that suggests code as they type, answers questions about code and can carry out coding tasks on its own.; it was offered through OpenAI's API until its retirement in 2023.
Origin
The 2021 model was introduced by Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan and colleagues in 'Evaluating Large Language Models Trained on Code' (July 2021). OpenAI revived the name for Codex CLI, an open-source terminal agent announced with the o3 and o4-mini models on 16 April 2025, and for the cloud agent on 16 May 2025.
Usage note
Because the name was reused, papers citing 'Codex' before 2025 mean the 2021 model, while news from 2025 onward usually means the agents.
Related
- coding agentcoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed.
- background agentbackground agent, n.: An AI agent you hand a task to and then leave alone: it works by itself in the cloud, often for many minutes or hours, and reports back with finished work for you to review.
- agent instruction fileagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session.
- GitHub CopilotGitHub Copilot, prop. n.: An AI assistant from GitHub, built into programmers' code editors, that suggests code as they type, answers questions about code and can carry out coding tasks on its own.
- HumanEvalHumanEval, prop. n.: A set of 164 small programming exercises, each described in plain English and checked by hidden tests, used to see whether an AI can write working Python code.
- Claude CodeClaude Code, prop. n.: A coding assistant from Anthropic that works as an agent: given a programming task, it explores the project, changes the code and runs programs to check its work, from the web, a desktop app, an editor or a terminal.
Sources
- Chen, M., Tworek, J., Jun, H., Yuan, Q. et al. (2021). Evaluating Large Language Models Trained on Code. arXiv:2107.03374. arxiv.org/abs/2107.03374
- OpenAI (2025). Introducing Codex. 16 May 2025. openai.com/index/introducing-codex
- OpenAI (2025). Introducing OpenAI o3 and o4-mini (announcing Codex CLI). 16 April 2025. openai.com/index/introducing-o3-and-o4-mini
Mentioned in
coding agentcoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed., GitHub CopilotGitHub Copilot, prop. n.: An AI assistant from GitHub, built into programmers' code editors, that suggests code as they type, answers questions about code and can carry out coding tasks on its own., OpenAIOpenAI, prop. n.: The American company that makes ChatGPT and the GPT models. It began in 2015 as a nonprofit research lab and is now a for-profit company overseen by a nonprofit foundation., agentic codingagentic coding, n.: Writing software by directing AI agents that plan, edit and test code themselves, with the human acting more as a reviewer and manager than as the person typing every line., AI coding assistantAI coding assistant, n.: AI software that helps people write computer programs — suggesting the next lines, explaining code, finding bugs or making changes on request — inside the tools programmers already use., background agentbackground agent, n.: An AI agent you hand a task to and then leave alone: it works by itself in the cloud, often for many minutes or hours, and reports back with finished work for you to review., code generationcode generation, n.: Having an AI write computer code from a description in plain language, from finishing a single line as you type to building whole programs., program-aided language modelprogram-aided language model, n.: A technique in which an AI solves a math or logic problem by writing a small computer program and letting a computer run it, instead of doing the arithmetic in its head.