Infinite Model response Hanging

Resolved 💬 3 comments Opened Nov 22, 2025 by rmakestrash-jpg Closed Nov 22, 2025
💡 Likely answer: A maintainer (github-actions[bot], contributor) responded on this thread — see the highlighted reply below.

What version of Codex is running?

v0.63.0

What subscription do you have?

Plus

Which model were you using?

Codex

What platform is your computer?

Windows 11

What issue are you seeing?

model never replys some kinda weird oath glitch stuck infite loading. it could be user error but idk.

What steps can reproduce the bug?

Uploaded thread: 019aad72-7048-74c3-b3f0-9209615edf52

What is the expected behavior?

PS C:\Users\14046\downloads> codex login
Starting local login server on http://localhost:1455.
If your browser did not open, navigate to this URL to authenticate:

https://auth.openai.com/oauth/authorize?response_type=code&client_id=app_EMoamEEZ73f0CkXaXp7hrann&redirect_uri=http%3A%2F%2Flocalhost%3A1455%2Fauth%2Fcallback&scope=openid%20profile%20email%20offline_access&code_challenge=jVSmyE9iZ0MApMptPIeStYwYCVtAJOclOufLiRkihaQ&code_challenge_method=S256&id_token_add_organizations=true&codex_cli_simplified_flow=true&state=91Rg7_SBWvC2c1xx5urLM05N7FrtijLkKJxP0whjuRc&originator=codex_cli_rs
/review - review any changes and find issues

› hi

■ Conversation interrupted - tell the model what to do differently. Something went wrong? Hit /feedback to report
the issue.
PS C:\Users\14046\downloads> code
PS C:\Users\14046\downloads> codex
╭─────────────────────────────────────────────────╮
│ >_ OpenAI Codex (v0.63.0) │
│ │
│ model: gpt-5.1-codex-max /model to change │
│ directory: ~\downloads │
╰─────────────────────────────────────────────────╯

To get started, describe a task or try one of these commands:

/init - create an AGENTS.md file with instructions for Codex
/status - show current session configuration
/approvals - choose what Codex can do without approval
/model - choose what model and reasoning effort to use
/review - review any changes and find issues

• Model changed to gpt-5.1 medium

› hi

• Feedback uploaded. Please open an issue using the following URL:

https://github.com/openai/codex/issues/new?template=2-bug-report.yml&steps=Uploaded%20thread:%20019aad72-7048-74c3-
b3f0-9209615edf52

Or mention your thread ID 019aad72-7048-74c3-b3f0-9209615edf52 in an existing issue.

◦ Working (7m 57s • esc to interrupt)

› Summarize recent commits

100% context left · ? for shortcuts

Additional information

_No response_

View original on GitHub ↗

3 Comments

github-actions[bot] contributor · 9 months ago

Potential duplicates detected. Please review them and close your issue if it is a duplicate.

  • #7182
  • #7181
  • #7179
  • #7178
  • #7177

Powered by Codex Action

etraut-openai contributor · 9 months ago

Thanks for reporting the problem. The service is down in some geographies. The team is working on it. Closing in favor of https://github.com/openai/codex/issues/7156.

ghost · 9 months ago

Title: Codex CLI rejects valid project API keys — "invalid_api_key" & Conversation interrupted

Description:
Hi — I’m hitting an auth failure when trying to run the Codex CLI with a project API key. I’ve followed the documented steps: removed ~/.codex/auth.json, set ~/.codex/config.toml with preferred_auth_method = "apikey", exported OPENAI_API_KEY in the shell, and use codex-cli v0.63.0. /status shows “API key configured (run codex login to use ChatGPT)”, but any attempt to talk to the model returns “Conversation interrupted” and curl requests to the API return:

{"error":{"message":"Incorrect API key provided: sk-proj-****CTED","code":"invalid_api_key"}}

What I tried:

  • Created >30 fresh Project API keys (sk-proj-...), exported them in the shell, and tested with curl — every key returns invalid_api_key.
  • Confirmed environment variable is set in the same terminal from which I launch Codex.
  • Deleted ~/.codex/auth.json and ~/.codex/sessions.
  • Updated Codex CLI to latest (v0.63.0).

Diagnostics (redact keys before posting):

  • codex --version: v0.63.0
  • /status block: [paste full /status output here]
  • ~/.codex/config.toml: [paste contents]
  • tail -n 200 ~/.codex/log: [paste the last ~200 lines]
  • curl output (redact the key): [paste curl response which contains invalid_api_key]

Problem summary:
Codex CLI appears to be sending a key that the backend rejects with invalid_api_key. The same keys created in the platform UI consistently return the invalid API key error when used from my machine. This prevents Codex from establishing a conversation (Conversation interrupted / Reconnecting). My account: ryan.lueckenotte@gmail.com.

Please advise whether:

  1. There’s an org/project restriction preventing project keys from working with Codex CLI, or
  2. There’s a bug in the CLI’s API-key authentication path that mangles keys or uses the wrong header, or
  3. Something in the backend is rejecting project keys generated for my account.

I can provide additional logs or live debugging info on request. Thanks.