OpenAI· Updated
OpenAI announced a performance roadmap for Codex that aims for a 10x speedup in 2026. This shift from incremental gains to an order-of-magnitude improvement signals a move toward near-instantaneous agentic execution for complex engineering tasks.
GPT-5.4 users. The team confirmed they have a clear engineering path to achieve this order-of-magnitude leap.Speed is the primary bottleneck for agentic workflows where models reason through multiple steps. This performance push builds on recent efforts to enable unattended engineering automations that execute long-running tasks without constant human intervention. A 10x improvement would move the user experience from waiting on an agent to real-time collaboration.
Currently, you can access the performance ceiling by using the /fast command in the Codex CLI. This roadmap extends OpenAI's focus on developer experience, following the release of tools for visualizing agent logs to help debug complex sessions. Learning the platform now prepares you for these upcoming speedups.
If you think Codex with GPT-5.4 is /fast already… we have line of sight for at least an order of magnitude in speedups this year. Good engineering compounds and it’s never been a better time to learn Codex.
Still wondering? A few quick answers below.
Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →
