Skip to content
IFAI

AI tools

OpenAI slashes ChatGPT and Codex desktop app storage by 60%

Published Reach 36

OpenAI just rolled out a disk storage update for its ChatGPT and Codex desktop apps, shrinking local state data by roughly 60%. The fix directly addresses feedback from power users who were rapidly running out of drive space. It is a small but welcome tune-up that hands precious gigabytes back to heavy users.

If you keep AI apps running all day while you code, your drive has probably felt the squeeze. OpenAI just rolled out a disk storage optimization for both the ChatGPT and Codex desktop apps, slashing their on-disk storage footprint by roughly 60%.

Announced by OpenAI team member thsottiaux as Day 9 of an ongoing rollout series, the change targets how the desktop applications serialize and write state to disk. The team described it as a small ship, but noted it was triggered by a steady stream of feedback from power users who were literally running out of disk space.

Cleaning up local bloat

Most people assume generative AI lives entirely in the cloud, but desktop clients quietly hoard conversation states, cached context, and local records over time.

For power users keeping these tools open across full workdays, that footprint grew fast. The new patch reworks that underlying storage mechanism directly, cutting stored state by well over half without changing how the apps actually feel to use.

Why it matters

For heavy users, especially developers running on storage-constrained laptops, reclaiming 60% of that disk footprint is an immediate quality-of-life win. You get the same daily assistant without watching your remaining gigabytes vanish.

We think this is a smart, grounded move. As AI assistants transition from occasional chat windows into permanent desktop companions, unglamorous plumbing like disk hygiene and memory management matters just as much as raw model intelligence.

Tackling local bloat head-on shows OpenAI is paying attention to the friction points of real-world, non-stop developer workflows rather than just shipping headline-grabbing features.

Sources (1)

  1. [1]
    x.com(opens in a new tab)Originalx.com

Similar events

Past happenings like this one, for context.

Best of AI, daily