Hi @JS-Prod,
That does look unusual.
Just to narrow it down: was gpt-5.5 xhigh explicitly pinned in the configuration for the affected tasks or was it selected only through the model picker?
Also, were any subagents involved? If so, do you know which models those subagents were configured to use? In some setups, subagents can run with different model and reasoning settings than the primary…
Show Codex a workflow once. Reuse it as a skill.
Record & Replay in Codex
Record & Replay lets you show Codex a recurring task, like filing an expense report or submitting a time-off request.
Codex turns that demo into an inspectable, editable skill.
You control when recording starts and stops.
More details in the docs:
developers.openai.com
IMHO: Nothing lost.
An attempt to force you into a UI to make a prompt as a reusable undiscoverable incomplete request shape that requires knowledge of it for success anyway.
Developed as playground-UI-only simply to increase the friction of simulating a “GPT” for a user on demand?
How to use the API like a normal person:
developers.openai.com
You can also create API keys using the Codex plugin for Developers.
developers.openai.com
Install the OpenAI Developers plugin in Codex to build, troubleshoot, and ship apps on top of the OpenAI API.
Or are you looking for something completely different?
It’s not so much the model, but more about how you prompt it - which, in my experience, takes trail and error. Prompting is an art.
Models change - they come and go.
I suggest you learn about prompt engineering. There are multiple OpenAI sources on the subject. Here is one:
developers.openai.com
Learn strategies and tactics for better results using large…
Hi @T.Maillet ,
Just to clarify: Sora 2 / Videos API is deprecated, but according to the docs it is scheduled to shut down on September 24, 2026. So a current 503 Service Unavailable error is not necessarily explained by the deprecation alone.
developers.openai.com
Learn how to generate, refine, and manage videos using the OpenAI Videos API.
Welcome to the community!
Yes Windows remote control support for Codex rolled out yesterday (May, 29), so it’s hard to tell from the outside whether this is a bug, rollout delay or an app/account state issue.
developers.openai.com
Latest updates to Codex, OpenAI’s coding agent
I’d first check that both the Windows Codex app and the ChatGPT mobile app…
Note that the knowlege cutoff date for GPT-5.5 is December 1, 2025
developers.openai.com
This topic has been discussed before. Foundation models are not very good at describing themselves. It appears that it hasn’t been fixed yet
Hi,
This may be a distinction between organization verification and the verification required for app submission.
The Apps SDK submission docs say that before submitting an app, you need to complete identity verification for the name you plan to publish under. If publishing under your own name, that means individual verification; if publishing under a business name, that means business…
Welcome to the community!
Since your screenshot shows both Individual and Business verification as approved, but the submission flow still says verification is required, I’d first check that you are submitting from the same organization where verification is approved.
developers.openai.com
Guidelines for submitting ChatGPT apps for…
I agree. The confusing part is that the Realtime API Beta was already deprecated on May 12, so you will need to update your code either way.
But this delay caught me by surprise as well.
Here is the deprecations page, which also includes a link to the migration guide.
developers.openai.com
As we launch safer and more capable…
Welcome to the forum!
Below each topic, you’ll find a Related topics section. Please search there first, as questions about sign-in and phones are common.
If you don’t find anything that helps, reply here to let us know you’re still having the problem. Please include what you tried and what didn’t work.
This may also be of value:
developers.openai.com
Serdjio:
And I’ve noticed something interesting: the forum admins carefully step around these complaints. No official statement, no “we’re working on it”, no “we’re looking for solutions.”
Where is this vague?
People avoid complaints like yours like the plague because the people bearing them usually just want to fight rather than face the facts.
Did anyone here on the plus plan not get a few…
Hi,
The docs say mobile setup requires the Codex App on a Mac host, signed into the same ChatGPT account and workspace as the phone. They also say the setup flow is not available from the Codex CLI or IDE extension.
developers.openai.com
Connect your phone or another device to a Codex host, or use projects on remote machines over SSH
So I…
I don’t use Android myself, but here’s something that I found and that might be worth checking, if you already haven’t that is.
Since the app works on browser/iOS and your server receives initialize, tools/list, resources/list and resources/read, but no tools/call ever reaches your server on Android, this sounds like the failure may be happening inside the ChatGPT Android client before…
Welcome to the forum!
First I am not an OpenAI employee.
I looked for a direct answer to that in the OpenAI documentation, FAQ, issues, etc. and by having ChatGPT search and we could not find a direct answer.
ChatGPT did note
developers.openai.com
Sign-in methods for Codex
and in paticular this statement
Codex cloud interacts directly with…
Hi,
I’d recommend to check this link:
developers.openai.com
Learn how to submit your app to the ChatGPT Apps Directory and Codex Plugin Directory.
Based on what I found on this document, I’d recommend to check the part that seems most relevant is the common rejection reason: “We’re unable to connect to your MCP server using…
Hi team,
I’m trying to set up the Codex Chrome extension using the official docs:
developers.openai.com
Use Chrome with Codex while managing browser permissions, website approvals, and browsing data
The docs say to open Codex, go to Plugins, and add the Chrome plugin. However, I don’t see any Chrome plugin in my plugin directory.
I…
Thanks for the link and the update on Docs. I think this one should also be noted as a stand-alone:
developers.openai.com
Control which commands Codex can run outside the sandbox
Hope this helps!
I edited the post to reference the auto-review mode a bit late. Not sure if you have seen it but that’s what I am using nowadays.
developers.openai.com
How Codex routes sandbox-boundary approvals through a reviewer agent
Quick and easy solution for everyday approvals.
Hi!
Thank you for raising this.
You are probably already aware that you can still use gpt-realtime-mini-2025-10-06. You do not need to use the undated model slug, which currently points to the December snapshot.
For production systems, dated model versions are usually the better choice because they help keep behavior more consistent.
Since I do not see a deprecation notice for the older…
内容に不足があったのでつくり直しました。
I rewrote this because important details were missing.
English
This is not a general complaint that “AI image editing is hard.”
This is not about whether the output looks visually similar.
This is not a criminal-law accusation.
This is about OpenAI’s ChatGPT Images 2.0 user-facing “editing” feature, and whether the product wording matches the observed…
It seems they have given up on ft for now, at least until we get some Stargates up and running with some gpu power to spare…
developers.openai.com
Find information about OpenAI API deprecations and recommended replacements.
Adding one more thing: are you checking the vector store file object specifically or the underlying file object?
From the docs, files can be removed from a vector store either by deleting the vector store file object or by deleting the underlying file object. Deleting the underlying file object removes it from all vector store and code interpreter configurations across assistants/threads in the…
Welcome to the forum!
My advice from seeing this would be to do exactly what the error message suggests: contact OpenAI Support and include the request ID.
OpenAI Support may also respond here if they see the thread, but I think opening a support ticket would still be useful, since this involves Responses API + webhooks, repeated server errors, slow response times and many request IDs. Forum…
Taragonn:
People may think I’m completely crazy for describing it this way. Maybe I am.
I would say you are correct.
Both OpenAI and Anthropic have been working hard in how to manage memory to get past problems of the context window size, learning from prompts by finding patterns and storing them, etc.
If one regularly reads the blogs, cookbooks, GitHub commits, etc., this becomes…
You might already know this, but from the Batch API docs, the completion window is currently set to 24h and batches that do not complete in time should eventually move to an expired state.
Docs:
developers.openai.com
Learn how to use OpenAI's Batch API for processing jobs with asynchronous requests, increased rate limits, and cost efficiency.
So if the batch is…
Welcome to the forum!
Besides what’s already been explained, I’m also adding these docs, so you can check if any of these might be of help.
Stored Chat Completions API reference:
OpenAI API Reference
Migration guide mentioning storage behavior for Responses / Chat Completions:
developers.openai.com
Data retention /…
Dirk_Peter:
After I prompted it to simply fix the “Forgot Password” function,
Dirk_Peter:
Let’s get one thing straight: The prompt was perfectly fine and precise – as always.
I’m not going to argue with you buddy based on what you feel is a vague prompt and what isn’t.
You’ve opened a thread about damage that is caused by a vague prompt, not by your definition, but by the definition of…
Hi and welcome to the community!
You are correct, the docs are outdated.
For API-specific questions, I generally suggest using developers.openai.com. Unfortunately, those docs also look incomplete here because they state that o-series reasoning models do not support stop sequences, which is not the full picture with regards to the gpt-5 family of models.
I will ping the team to review these…