codex doesn't respond in chat

Resolved 💬 8 comments Opened Jul 17, 2025 by laphang Closed Jul 29, 2025
💡 Likely answer: A maintainer (etraut-openai, contributor) responded on this thread — see the highlighted reply below.

What version of Codex is running?

codex-cli 0.7.0

Which model were you using?

o3 via azure

What platform is your computer?

Linux 5.10.237-230.949.amzn2.x86_64 x86_64 x86_64 (sagemaker studio terminal)

What steps can reproduce the bug?

run codex, and then just ask it a simple question (eg "hello, are you there?"), the llm call finishes but it doesn't reply in chat. It consistently does not reply in the chatbox.

(see section below for the output)

it is able to do bash operations and write code, and actually if you ask it a question and ask it to output the answer to a file, it will do that too.

I get the same behaviour (not replying) in v0.4.0 as well.

What is the expected behavior?

It should be able to reply. Eg working below with v0.1.2505172129

● OpenAI Codex (research preview) v0.1.2505172129            │
╰──────────────────────────────────────────────────────────────╯
╭──────────────────────────────────────────────────────────────╮
│ localhost session: 12ed4f14b9654b82b58b840b5b7f4c18          │
│ ↳ workdir: ~/.../reasoning_scripts/gemini-iac                │
│ ↳ model: MLAI-o3                                             │
│ ↳ provider: azure                                            │
│ ↳ approval: suggest                                          │
╰──────────────────────────────────────────────────────────────╯
user
hello, are you there?

codex
Hello! I’m here and ready to help. What can I do for you today?

What do you see instead?

╭Messages (tab to focus)──────────────────────────────────────────────────────────────────────────────────────────────╮
│OpenAI Codex v0.7.0 (research preview)                                                                               │
│                                                                                                                     │
│codex session f15f1f14-a72c-4032-9c3b-8970cc6ff3e3                                                                   │
│workdir:                                                                                                             │
│/mnt/custom-file-systems/efs/fs-0f0f5d2fb048988d2_fsap-02598ae65ca0a1c25/mlai-ds-dev/mlai-prompt-layer-training/src/ │
│llm/analysis/reasoning_scripts/gemini-iac                                                                            │
│model: MLAI-o3                                                                                                       │
│provider: azure                                                                                                      │
│approval: UnlessTrusted                                                                                              │
│sandbox: read-only                                                                                                   │
│                                                                                                                     │
│user                                                                                                                 │
│hello, are you there?                                                                                                │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
│                                                                                                                     │
╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯
╭─────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
│ send a message — 1431 tokens used                                                                                   │
╰───────────────────────────────────────

Additional information

_No response_

View original on GitHub ↗

8 Comments

vitatea2023 · 1 year ago

run it with RUST_LOG=debug and then tail -F ~/.codex/log/codex-tui.log to see if that tells you anything interesting.

laphang · 1 year ago

Thanks for the reply, I get this below:

ng_scripts/gemini-iac$ RUST_LOG=debug codex --profile o3
sagemaker-user@default:~/user-default-efs/mlai-ds-dev/mlai-prompt-layer-training/src/llm/analysis/reasoning_scripts/gemini-iac$ tail -F ~/.codex/log/codex-tui.log
2025-07-17T06:08:21.949206Z DEBUG Submission sub=Submission { id: "2", op: AddToHistory { text: "hello, are you there?" } }
2025-07-17T06:08:21.949244Z DEBUG Recording items for conversation: [Message { role: "user", content: [InputText { text: "hello, are you there?" }] }]
2025-07-17T06:08:21.949263Z DEBUG Recording items for conversation: []
2025-07-17T06:08:21.949530Z DEBUG starting new connection: https://openai-digitalproduct-dev-01-eastus2.openai.azure.com/    
2025-07-17T06:08:21.952460Z DEBUG connecting to 20.62.58.5:443
2025-07-17T06:08:22.145435Z DEBUG connected to 20.62.58.5:443
2025-07-17T06:08:23.440760Z DEBUG sse event other="response.content_part.added"
2025-07-17T06:08:24.131730Z DEBUG pooling idle connection for ("https", openai-digitalproduct-dev-01-eastus2.openai.azure.com)
2025-07-17T06:08:24.131736Z DEBUG Output item item=Reasoning { id: "rs_687893572bd0819d960f56186a1e78f40bb98603cef21201", summary: [] }
2025-07-17T06:08:24.131855Z DEBUG Turn completed
laphang · 1 year ago

thanks, will wait for the next release

laphang · 11 months ago

FYI this was resolved in v0.8.0

henrypost · 8 months ago

It looks like this issue is still occurring in 0.58.0:

Windows PowerShell
Copyright (C) Microsoft Corporation. All rights reserved.
(base) PS C:\Users\henrypost> npx @openai/codex@latest


› test


› test



› test
/quit

(base) PS C:\Users\henrypost> ^C
(base) PS C:\Users\henrypost> ^C
(base) PS C:\Users\henrypost> npm update @openai/codex

up to date in 2s

(base) PS C:\Users\henrypost> npm view  @openai/codex version
0.58.0
etraut-openai contributor · 8 months ago

@henrypost, this issue is closed. If you think that you're seeing a similar issue with a recent version of codex, please open a new issue and provide details using the /feedback slash command.

henrypost · 8 months ago

TY @etraut-openai - using latest version works!!