
Share
OpenAI's new GPT-5-Codex enhances Codex with advanced training data and performance tweaks, making it an indispensable tool for developers tackling complex coding challenges.
OpenAI has announced significant upgrades to Codex, their AI-powered coding assistant. The latest version, dubbed GPT-5-Codex, is optimized for real-world software engineering tasks, offering improved performance, reliability, and seamless integration across various development environments.
GPT-5-Codex is a specialized variant of OpenAI’s GPT-5 model, fine-tuned to excel in coding scenarios. Here are the key technical improvements:
Training Data: The model was trained on a diverse set of real-world software engineering tasks, including:
Performance Enhancements:
Integration:
For developers, these upgrades translate into a more effective coding collaborator that can:

To start using GPT-5-Codex, you can install the Codex CLI globally via npm:
$ npm i -g @openai/codex
If you’re already a Codex user, GPT-5-Codex is available through your ChatGPT subscription. It’s included with ChatGPT Plus, Pro, Business, Edu, and Enterprise plans.
Two weeks ago, OpenAI unified Codex into a single product experience connected by your ChatGPT account. This update enables you to move work seamlessly between your local environment and the cloud. Here are some highlights:
As of September 23, 2025, GPT-5-Codex is available to developers using Codex via API key. It’s priced the same as GPT-5 and can be accessed through the Responses API. Regular updates to the underlying model snapshot will ensure you always have access to the latest improvements.
For more details, check out the Codex developer documentation and changelog.
With these enhancements, Codex is moving closer to becoming a reliable teammate that understands your context, works alongside you, and takes on tasks for your team.
Tags
Original Sources
About the author
Kai built ML infrastructure at a Bay Area startup before developing an obsession with transformer architectures and inference optimisation that eventually pulled him out of product work entirely. A stint at a compute research lab sharpened his instinct for what actually matters in a model release versus what is marketing. He writes from the inside — from the perspective of someone who has debugged the systems he is describing at three in the morning. He is allergic to hype and instinctively drawn to the unglamorous plumbing questions that everyone else skips over.
More from The Engineer →This Week's Edition
16 September 2025
133 articles
Related Articles

Smarter Engagement for Stronger Growth: How Payers Can Leverage AI to Do More with Less
Products & Applications · 3 min

Penn Medicine and K Health Deploy AI Clinical Agents to Enhance Patient Care
Products & Applications · 3 min

Wheel and b.well Partner to Build Turnkey AI-First Virtual Care Infrastructure
Products & Applications · 3 min
Related Articles

Smarter Engagement for Stronger Growth: How Payers Can Leverage AI to Do More with Less
Products & Applications · 3 min

Penn Medicine and K Health Deploy AI Clinical Agents to Enhance Patient Care
Products & Applications · 3 min

Wheel and b.well Partner to Build Turnkey AI-First Virtual Care Infrastructure
Products & Applications · 3 min
More Stories