OpenAI's Codex app is now available on macOS – and it’s free for some ChatGPT users for a limited time
OpenAI has rolled out the macOS app to help developers make more use of Codex in their work
Sign up today and you will receive a free copy of our Future Focus 2025 report - the leading guidance on AI, cybersecurity and other IT challenges as per 700+ senior executives
You are now subscribed
Your newsletter sign-up was successful
OpenAI wants more people to code with Codex, so it’s rolling out an app for macOS and making the AI coding tool available to free users for a limited time.
The Codex app for macOS lets coders run multiple AI agents at once and work with agents over long-running tasks, the company explained in a blog post.
OpenAI first unveiled Codex in April 2025, updating it with new features in September and unveiling a fresh model just weeks ago.
The release comes amid competition from rival AI developers, in particular GitHub Copilot, Anthropic Claude Code, and Google's recently released free coding agent Jules.
"The Codex app changes how software gets built and who can build it — from pairing with a single coding agent on targeted edits to supervising coordinated teams of agents across the full lifecycle of designing, building, shipping, and maintaining software," the company noted.
OpenAI said its models were now able to manage complex, long-running tasks, meaning developers are now searching for ways to organise those AI efforts.
"The core challenge has shifted from what agents can do to how people can direct, supervise, and collaborate with them at scale — existing IDEs and terminal-based tools are not built to support this way of working," OpenAI added.
Sign up today and you will receive a free copy of our Future Focus 2025 report - the leading guidance on AI, cybersecurity and other IT challenges as per 700+ senior executives
What users can expect with Codex
Codex is designed to act as a “command center for agents”, according to OpenAI, helping streamline multi-tasking across different projects and allowing users to review agents’ work.
"It also includes built-in support for worktrees, so multiple agents can work on the same repo without conflicts," OpenAI explained.
"Each agent works on an isolated copy of your code, allowing you to explore different paths without needing to track how they impact your codebase. As an agent works, you can check out changes locally or let it continue making progress without touching your local git state."
The company stressed that the system was built with security in mind, with agents only allowed to edit files in the designated folder or branch, and asking for permission to run commands that require elevated permissions.
OpenAI said Codex will generally be available via higher-end, paid-for subscriptions, including ChatGPT Plus, Pro, Business, Enterprise and Edu.
However, for a limited time it will also offer Codex via ChatGPT Free and Go in order to "show more people what's now possible".
OpenAI added that it would double rate limits on paid-for plans for a limited time too.
Beyond coding
The Codex app can do more than manage code generation, according to OpenAI. Using its previously announced "skills" the tool can gather information, solve problems, write and more, all controlled via the new app.
OpenAI said it had developed hundreds of skills to help development teams, ranging from triaging bugs to deploying to the cloud and creating documents.
"You can explicitly ask Codex to use specific skills, or let it automatically use them based on the task at hand," OpenAI noted.
For example, OpenAI said it used Codex to develop a racing game. Beyond code development and testing – which it did by playing the game – they also used an image generation skill and a web game development skill.
The game was created using one single – but complex and detailed – prompt, with a selection of follow-up prompts to finish the work.
These skills can be automated and scheduled, in particular to run repetitive tasks such as triaging daily issues, checking for bugs and more.
Pick a personality
After the release of GPT-5, OpenAI had to walk back changes to the "personality" of its chatbot – and it clearly heard that feedback.
For Codex, developers will be able to choose a "personality" they prefer to work with on a daily basis.
However, there's only two to start: a blunt, terse and pragmatic style, or a "more conversational, empathetic one".
FOLLOW US ON SOCIAL MEDIA
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
You can also follow ITPro on LinkedIn, X, Facebook, and BlueSky.
Freelance journalist Nicole Kobie first started writing for ITPro in 2007, with bylines in New Scientist, Wired, PC Pro and many more.
Nicole the author of a book about the history of technology, The Long History of the Future.
-
Microsoft has a new AI poster child in Anthropic – and it’s about timeOpinion Microsoft is cosying up to Anthropic at a crucial time in the race to deliver on AI promises
-
Will AI hiring entrench gender bias?ITPro Podcast This International Women's Day, it's more important than ever to consider the inherent biases of training data
-
Why Amazon’s ‘go build it’ AI strategy aligns with OpenAI’s big enterprise pushNews OpenAI and Amazon are both vying to offer customers DIY-style AI development services
-
Salesforce targets telco gains with new agentic AI toolsNews Telecoms operators can draw on an array of pre-built agents to automate and streamline tasks
-
February rundown: SaaS-pocalypse now?ITPro Podcast Geopolitical uncertainty is intensifying public and private sector focus on true sovereign workloads
-
‘A huge vote of confidence’: London set to host OpenAI's largest research hub outside USNews OpenAI wants to capitalize on the UK’s “world-class” talent in areas such as machine learning
-
Sam Altman just said what everyone is thinking about AI layoffsNews AI layoff claims are overblown and increasingly used as an excuse for “traditional drivers” when implementing job cuts
-
Kyndryl wants to help enterprises keep AI agents in line – and avoid costly compliance blundersNews Controls become machine‑readable policies that AI agents can read and must obey


