OpenAI has released "a major update" to Codex which it says will help make the platform a more effective workplace tool for users.
Codex will now be able to go "beyond coding" and access other parts of your computer, as well as operating desktop apps by itself, running in the background so it doesn't interfere with your current work.
And thanks to new automation tools, Codex will also now be able to remember from past experiences but also schedule future work for itself, "waking up" automatically to continue on a long-term task, even across several days or weeks.
A smarter Codex for all
"We’re releasing a major update to Codex, making it a more powerful partner for the more than 3 million developers who use it every week to accelerate work across the full software development lifecycle," OpenAI said in a blog post announcing the news.
"Codex can now operate your computer alongside you, work with more of the tools and apps you use everyday, generate images, remember your preferences, learn from previous actions, and take on ongoing and repeatable work."
The new memory function, now in preview, will mean Codex can remember useful context from previous experiences, such as personal preferences, corrections and information that took time to gather, helping complete future tasks faster and more effectively. It will also proactively proposes useful work to continue where you left off using context from projects, connected plugins, and memory.
This increased capability means multiple agents can now work together on a Mac at the same time, but avoid interfering with your work in other apps - a tool OpenAI notes could be useful for developers iterating on frontend changes, testing apps, or working in apps that don’t expose an API.
Codex now also includes an in-app browser, allowing users to comment directly on pages to provide precise instructions to the agent, which again could be useful for frontend and game development.
As for image generation, Codex will be able to use gpt-image-1.5 to create visuals quickly, which combined with screenshots and code, will help workers create product concepts, frontend designs, mockups, and games inside the same workflow.
Elsewhere, there's also support for addressing GitHub review comments, running multiple terminal tabs, and users will also be able to open files directly in the sidebar with rich previews for PDFs, spreadsheets, slides, and docs, and use a new summary pane to track agent plans, sources, and artifacts.
All the updates are rolling out to Codex desktop app users signed in with ChatGPT, with personalization features (including context-aware suggestions and memory) available for Enterprise, Edu, and EU and UK users soon, and computer use initially available on macOS, before rolling out to EU and UK users soon.
