GPT-5.3 Codex in IntelliJ IDEA is extremely slow even with 32GB memory allocated

Answered
Avatar
Sg Yang
Created

Hello,

I'm using GPT-5.3 Codex in IntelliJ IDEA, but it's extremely slow and the response time is very long.

I already increased the IDE memory allocation to 32GB via Help → Change Memory Settings, but the performance is still very poor.

I have a few questions:

  • Is anyone else experiencing the same issue?
  • Is this a known issue when using GPT-5.3 Codex in IntelliJ IDEA?
  • Are there any recommended settings or optimizations to improve performance?

Any help would be greatly appreciated. Thank you.

Votes

0

Share

1 comment
Avatar
Gleb Gladcov
Created

Hello,

Yes, we do have a known issue where Codex can cause IDE slowdowns/freezes in certain scenarios. 
It is currently tracked here:
LLM-24887 IDE freezes / high CPU when Codex (Agent mode) executes PowerShell commands and renders large agent output in AI Chat

Could you please review the ticket and its suggested workaround to see if it matches your case?

If it does not cover your scenario, please open a Support ticket so we can investigate further. 
When submitting it, please include a log archive collected with the following debug settings enabled:

  1. Navigate to Help → Diagnostic Tools → Debug Log Settings
  2. Add:
    com.intellij.ml.llm
    com.intellij.util.io.HttpRequests
    com.intellij.util.proxy.CommonProxy
    com.intellij.ml.llm.agents.codex
  3. Restart the IDE
  4. Reproduce the issue
  5. Collect logs via Help → Collect Logs and Diagnostic Data

You can create a Support request using this link: Support portal.

0

Please sign in to leave a comment.