Environment

• Windows 11 x64 (build 26200)
• OpenAI Codex 26.707.9981.0 x64
• NVIDIA GeForce RTX 4060 Laptop GPU
• No SecureLink or third-party DLL injectors installed
• App was stable for several weeks before July 15

Behaviour
The app does not close — instead, ChatGPT.exe silently crashes in the
background 6+ times per minute. Each crash freezes the UI for 2–5
seconds. Clicking a conversation, typing, switching tabs — all stall
intermittently. The freezes repeat in a loop throughout normal use.

Over 24 hours this produced 30+ crash dumps (12 MB each, ~360 MB
total disk writes), making the UI lag even worse.

Crash details

• Exception: 0xC06D007F (delay-load “procedure not found” failure)
• Faulting module: @serialport/bindings-c
• Windows Error Reporting confirms identical fault offset across all 30+ dumps

Troubleshooting attempted (no change)

• Clean .codex directory, archived 200+ conversations, disabled 9 of 14 plugins
• Installed latest VC++ 2015–2022 Redistributable (14.44)
• Disabled GPU acceleration, cleared all caches, reinstalled and reset the app
• VCRUNTIME140_1.dll is present and up to date
• Suppressed LocalDumps for ChatGPT.exe (stopped the I/O stall, did not stop the crash)

Summary
This is the same 0xC06D007F / serialport.node / device-kit-oai crash
reported in GitHub Issue #33381 and community post #1387026, but on
x64 the symptom is intermittent UI freezing rather than a hard crash on
startup. The x64 ChatGPT.exe does export napi_* symbols, so the root
cause on x64 may be a prebuilt binary dependency mismatch rather than
missing NAPI exports — but the crash chain is identical. Could OpenAI
please either disable the CodexMicro feature flag for affected x64
builds, or ship a corrected win32-x64 serialport native module?

Translated by AI:

After the restart at 16:02, ChatGPT.exe, the main Codex process, silently crashed more than eight times within three minutes, always with exception code 0xc06d007f. Between 16:05 and 16:07, the entire application also crashed and restarted on its own. The UI froze for several seconds each time.

Both the old version (26.707.9981) and today’s updated version (26.707.12708) crash in the same way, so the update did not fix it.

Why this is clearly an official bug:

Windows 11 build 26200, the same version, worked normally until July 15. ChatGPT.exe now crashes more than six times per minute, freezing for 2–5 seconds each time.

There are multiple similar issues on GitHub. Several specifically mention the Windows 11 25H2 preview build 26200, which is exactly your system version: crashing whenever a message is sent, crashing in a loop immediately after launch, and freezing when opening the built-in browser.

• 刚才 16:02 重启后,ChatGPT.exe（Codex 的主程序）在 3 分钟内静默崩溃了 8 次以上，异常码都是 0xc06d007f；16:05–16:07
  之间整个应用还自己崩溃重启了一轮。每次崩溃 UI 就冻住几秒。
• 旧版本（26.707.9981）和今天更新的新版本（26.707.12708）都一样崩，说明升级救不了。

为什么确定是官方 bug：

• Windows 11
  build 26200、同版本、7 月 15 日之前一直正常、ChatGPT.exe 每分钟崩 6 次以上、每次卡 2-5 秒。
• GitHub 上有多个同类 issue，好几个都点名 Windows 11 25H2 预览版 build 26200（正是你的系统版本）：发任何消息就崩
  、启动即崩循环
  、打开内置浏览器卡死
  。

The Windows desktop application (ChatGPT/Codex) continuously freezes and enters a “Not Responding” state immediately after launching. It frequently triggers Windows “Efficiency Mode” automatically, and it gets stuck in a loop during the “Downloading” state. Clearing cache, using powercfg commands, or running as administrator does not fix the issue. This appears to be a global optimization bug with the latest update. Please investigate this for Windows users as the app is currently completely unusable.

For those new to the forum, welcome to the forum!

You are welcome to discuss Codex issues on the forum, however the official place to report and track them is the GitHub issue tracker for OpenAI Codex.

I had ChatGPT look for the closest related issue and it identified:

An OpenAI staff developer noted

Hi all - thanks a lot for the reports and we have disabled the feature for windows at the moment. The config itself might take some time to propagate and would definitely recommend a restart before it takes effect. Please let us know if you are still seeing the issue persistently after restarting in the next couple hours. We are also preparing a proper fix and it is currently in the pipeline.

If that issue matches your problem, consider adding a reaction on the GitHub issue page itself, not just on the forum topic, as reactions on GitHub help the developers gauge impact and prioritize work.

If it is not the same issue, try searching the existing GitHub issues for a closer match. If you cannot find one, then open a new issue with as much detail as possible.