# \#codex **URL:** https://community.openai.com/tag/codex/6.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [Warn users when they are close to their shared usage limit](https://community.openai.com/t/warn-users-when-they-are-close-to-their-shared-usage-limit/1407953)
**Author:** [@12k](https://community.openai.com/u/12k)\ **Replies:** 0\ **Last updated:** [October 20, 2026, 3:46am UTC](https://community.openai.com/t/warn-users-when-they-are-close-to-their-shared-usage-limit/1407953 "2026-10-20T03:46:20Z")
I’d like a warning when my remaining usage is getting low, for example when it reaches 15%. Ideally, the warning threshold would be configurable. I use Codex heavily for coding, but I also use ChatGPT for other things. … --- ## [Title: Feature request: Let a dot connect to multiple personal computers at the same time](https://community.openai.com/t/title-feature-request-let-a-dot-connect-to-multiple-personal-computers-at-the-same-time/1407934)
**Author:** [@DIO](https://community.openai.com/u/DIO)\ **Replies:** 0\ **Last updated:** [October 20, 2026, 2:57am UTC](https://community.openai.com/t/title-feature-request-let-a-dot-connect-to-multiple-personal-computers-at-the-same-time/1407934 "2026-10-20T02:57:47Z")
I use a Windows computer at work and a Mac at home, with the Codex desktop app signed in on both. I want my dot to help me follow up on projects across both machines. In my case, after connecting the Mac, the Windows PC… --- ## [Cooking with Codex: Day 15 of 28 days of Quality of Life improvements and/or a full Reset](https://community.openai.com/t/cooking-with-codex-day-15-of-28-days-of-quality-of-life-improvements-and-or-a-full-reset/1403525)
**Author:** [@VeitB](https://community.openai.com/u/VeitB)\ **Replies:** 71\ **Last updated:** [October 20, 2026, 2:00am UTC](https://community.openai.com/t/cooking-with-codex-day-15-of-28-days-of-quality-of-life-improvements-and-or-a-full-reset/1403525 "2026-10-20T02:00:45Z")
It looks like the OpenAI team decided that the flood of DevDay announcements was not enough. :grinning\_face\_with\_smiling\_eyes: Starting today, there will be one clear improvement each day for most of all Codex or Work u… --- ## [Has anyone moved existing Codex chats between Windows PCs with their full history?](https://community.openai.com/t/has-anyone-moved-existing-codex-chats-between-windows-pcs-with-their-full-history/1407878)
**Author:** [@MicrobesToModules](https://community.openai.com/u/MicrobesToModules)\ **Replies:** 2\ **Last updated:** [October 20, 2026, 1:02am UTC](https://community.openai.com/t/has-anyone-moved-existing-codex-chats-between-windows-pcs-with-their-full-history/1407878 "2026-10-20T01:02:02Z")
I’m moving my Codex work between two Windows computers and want to keep using two existing chats with their full history. Has anyone successfully moved existing Codex desktop chats between Windows PCs and continued them… --- ## [Feature request: clearer Codex usage and credit dashboard](https://community.openai.com/t/feature-request-clearer-codex-usage-and-credit-dashboard/1397605)
**Author:** [@0uriel.00](https://community.openai.com/u/0uriel.00)\ **Replies:** 6\ **Last updated:** [October 20, 2026, 1:01am UTC](https://community.openai.com/t/feature-request-clearer-codex-usage-and-credit-dashboard/1397605 "2026-10-20T01:01:12Z")
Codex currently shows several separate limits: a 5-hour limit, a weekly limit, plan-included usage, and purchased credits. It is difficult to understand how these limits interact and when purchased credits begin to be us… --- ## [I can’t trust Codex with long tasks anymore. It keeps moving the goalposts](https://community.openai.com/t/i-can-t-trust-codex-with-long-tasks-anymore-it-keeps-moving-the-goalposts/1407513)
**Author:** [@Nigel999](https://community.openai.com/u/Nigel999)\ **Replies:** 4\ **Last updated:** [October 20, 2026, 12:17am UTC](https://community.openai.com/t/i-can-t-trust-codex-with-long-tasks-anymore-it-keeps-moving-the-goalposts/1407513 "2026-10-20T00:17:26Z")
I’m running into the same problem over and over when using Codex for longer coding tasks: it gradually loses track of the original goal, expands the scope on its own, and starts working on things I never asked for. I ca… --- ## [OpenAI fix for GPT-6 Luna/Sol image understanding](https://community.openai.com/t/openai-fix-for-gpt-6-luna-sol-image-understanding/1401333)
**Author:** [@platypus](https://community.openai.com/u/platypus)\ **Replies:** 1\ **Last updated:** [October 19, 2026, 11:19pm UTC](https://community.openai.com/t/openai-fix-for-gpt-6-luna-sol-image-understanding/1401333 "2026-10-19T23:19:37Z")
According to OpenAI devs, they’ve fixed a bug that led to degradation in image understanding on GPT-6 Luna and Sol. The improvements should be visible on both API and Codex/ChatGPT Desktop APP, including also computer us… --- ## [GPT-5.6 struggles with context drift and state loss](https://community.openai.com/t/gpt-5-6-struggles-with-context-drift-and-state-loss/1401327)
**Author:** [@reynoldsabernathy](https://community.openai.com/u/reynoldsabernathy)\ **Replies:** 7\ **Last updated:** [October 19, 2026, 6:19pm UTC](https://community.openai.com/t/gpt-5-6-struggles-with-context-drift-and-state-loss/1401327 "2026-10-19T18:19:23Z")
While GPT-5.6 had already lost some of its sharpness starting in late August, it has become even dumber in the last week. This isn’t a case of being routed to a “mini” model that answers without thinking—my account hasn’… --- ## [Codex parent-thread cache behavior while waiting on delegated tasks](https://community.openai.com/t/codex-parent-thread-cache-behavior-while-waiting-on-delegated-tasks/1391915)
**Author:** [@Brynjolm](https://community.openai.com/u/Brynjolm)\ **Replies:** 1\ **Last updated:** [October 19, 2026, 6:06pm UTC](https://community.openai.com/t/codex-parent-thread-cache-behavior-while-waiting-on-delegated-tasks/1391915 "2026-10-19T18:06:16Z")
I’m trying to understand the prompt-cache behavior of a long-lived Codex coordination thread that delegates work to separate threads and then remains idle while those delegated tasks continue executing. Specifically, as… --- ## [Feature Request: Optional Handoffs and Orchestration Across Chat, Work and Codex](https://community.openai.com/t/feature-request-optional-handoffs-and-orchestration-across-chat-work-and-codex/1395179)
**Author:** [@PsyHaIceFire](https://community.openai.com/u/PsyHaIceFire)\ **Replies:** 1\ **Last updated:** [October 19, 2026, 2:23pm UTC](https://community.openai.com/t/feature-request-optional-handoffs-and-orchestration-across-chat-work-and-codex/1395179 "2026-10-19T14:23:20Z")
Several existing feature requests propose individual parts of this idea: switching between Chat and Work while preserving context, better ChatGPT↔Codex handoffs, or automatic routing between environments. I would like t… --- ## [Codex repeatedly claims tasks are complete, then discovers major failures. Why are users paying for this?](https://community.openai.com/t/codex-repeatedly-claims-tasks-are-complete-then-discovers-major-failures-why-are-users-paying-for-this/1407206)
**Author:** [@anburira](https://community.openai.com/u/anburira)\ **Replies:** 3\ **Last updated:** [October 19, 2026, 11:36am UTC](https://community.openai.com/t/codex-repeatedly-claims-tasks-are-complete-then-discovers-major-failures-why-are-users-paying-for-this/1407206 "2026-10-19T11:36:43Z")
Codex failed six consecutive codebase audits. Why are users paying for the model’s repeated failures? I want to raise a serious issue with Codex’s reliability, its repeated claims of completion, and the financial burden … --- ## [Improving GPT's capacity for human-like reasoning is an urgent necessity](https://community.openai.com/t/improving-gpts-capacity-for-human-like-reasoning-is-an-urgent-necessity/1401828)
**Author:** [@mgm2](https://community.openai.com/u/mgm2)\ **Replies:** 1\ **Last updated:** [October 18, 2026, 11:23pm UTC](https://community.openai.com/t/improving-gpts-capacity-for-human-like-reasoning-is-an-urgent-necessity/1401828 "2026-10-18T23:23:53Z")
In the GPT-4 era, the model excelled at nuanced observation and expression. With the advent of GPT-5, the focus shifted to cost-efficiency and logical reasoning capabilities; while this makes sense, it has been taken to… --- ## [ChatGPT ↔ Codex Bridge: A Bidirectional Workflow for Software Architecture and Development](https://community.openai.com/t/chatgpt-codex-bridge-a-bidirectional-workflow-for-software-architecture-and-development/1407221)
**Author:** [@julianovmartins](https://community.openai.com/u/julianovmartins)\ **Replies:** 1\ **Last updated:** [October 18, 2026, 10:20pm UTC](https://community.openai.com/t/chatgpt-codex-bridge-a-bidirectional-workflow-for-software-architecture-and-development/1407221 "2026-10-18T22:20:47Z")
Feature Request: ChatGPT ↔ Codex Bridge Dear OpenAI Team, I would like to propose a native integration between regular ChatGPT conversations and Codex, with a simple action to generate a prompt and send it directly to C… --- ## [Windows Codex Desktop — Prompt Submission Hangs Indefinitely After First Working Chat](https://community.openai.com/t/windows-codex-desktop-prompt-submission-hangs-indefinitely-after-first-working-chat/1392974)
**Author:** [@mmarahman4847](https://community.openai.com/u/mmarahman4847)\ **Replies:** 10\ **Last updated:** [October 18, 2026, 7:19pm UTC](https://community.openai.com/t/windows-codex-desktop-prompt-submission-hangs-indefinitely-after-first-working-chat/1392974 "2026-10-18T19:19:45Z")
Codex Desktop on Windows becomes unable to send prompts. After submitting a message, the UI enters the running/loading state indefinitely and no response is returned. The issue appeared after updating the desktop applic… --- ## [Passed Open Source Project but lost email](https://community.openai.com/t/passed-open-source-project-but-lost-email/1393724)
**Author:** [@LysanderZ](https://community.openai.com/u/LysanderZ)\ **Replies:** 1\ **Last updated:** [October 18, 2026, 7:02pm UTC](https://community.openai.com/t/passed-open-source-project-but-lost-email/1393724 "2026-10-18T19:02:54Z")
I previously successfully applied for codex for Open Source Project and was approved. Thank you to the codex and openai teams! Unfortunately, due to my job change and my own negligence, I lost the original email. Now I … --- ## [You can just build things (with hardware)](https://community.openai.com/t/you-can-just-build-things-with-hardware/1407523)
**Author:** [@platypus](https://community.openai.com/u/platypus)\ **Replies:** 1\ **Last updated:** [October 18, 2026, 6:42pm UTC](https://community.openai.com/t/you-can-just-build-things-with-hardware/1407523 "2026-10-18T18:42:47Z")
There are some really cool hardware projects you can build with Codex. OpenAI maintains a website where you can learn, share and vote on various projects: https://youcanjustbuildthings.openai.chatgpt.site/ I have an old… --- ## [GPT-6.1 Sol suddenly consuming much more Plus usage since October 16](https://community.openai.com/t/gpt-6-1-sol-suddenly-consuming-much-more-plus-usage-since-october-16/1407305)
**Author:** [@SemperFi](https://community.openai.com/u/SemperFi)\ **Replies:** 12\ **Last updated:** [October 18, 2026, 4:55pm UTC](https://community.openai.com/t/gpt-6-1-sol-suddenly-consuming-much-more-plus-usage-since-october-16/1407305 "2026-10-18T16:55:05Z")
I’m a ChatGPT Plus subscriber using GPT-6.1 Sol in Work. Starting yesterday, October 16, I noticed a dramatic increase in how quickly it consumes my usage allowance. Fast mode is OFF. Previously, Sol 6.1 gave me a good… --- ## [Preserve the native GPT-6 Chat conversational experience within ChatGPT Work and Codex](https://community.openai.com/t/preserve-the-native-gpt-6-chat-conversational-experience-within-chatgpt-work-and-codex/1406842)
**Author:** [@ausgamer23](https://community.openai.com/u/ausgamer23)\ **Replies:** 1\ **Last updated:** [October 18, 2026, 2:23pm UTC](https://community.openai.com/t/preserve-the-native-gpt-6-chat-conversational-experience-within-chatgpt-work-and-codex/1406842 "2026-10-18T14:23:34Z")
I would like OpenAI to make the conversational behaviour of GPT-6 Instant, as experienced in ordinary ChatGPT, available within ChatGPT Work and potentially Codex. This is not simply a request for GPT-6 model availabili… --- ## [Feature request: “One continuous assistant” across ChatGPT + Work + Codex + Dots (shared, user-controlled continuity)](https://community.openai.com/t/feature-request-one-continuous-assistant-across-chatgpt-work-codex-dots-shared-user-controlled-continuity/1407372)
**Author:** [@jjm149](https://community.openai.com/u/jjm149)\ **Replies:** 0\ **Last updated:** [October 18, 2026, 3:13am UTC](https://community.openai.com/t/feature-request-one-continuous-assistant-across-chatgpt-work-codex-dots-shared-user-controlled-continuity/1407372 "2026-10-18T03:13:07Z")
TL;DR I’d like ChatGPT, Work, Codex, and Dots to feel like one trusted assistant with a consistent personality and shared understanding of my goals/preferences—with explicit user control over what carries across each su… --- ## [GUI: Allow dunder in codex prompt window](https://community.openai.com/t/gui-allow-dunder-in-codex-prompt-window/1407365)
**Author:** [@alexander.podgorny](https://community.openai.com/u/alexander.podgorny)\ **Replies:** 0\ **Last updated:** [October 18, 2026, 2:07am UTC](https://community.openai.com/t/gui-allow-dunder-in-codex-prompt-window/1407365 "2026-10-18T02:07:56Z")
When typed “\_\_something\_\_”, codex prompt window turns it into bold. This makes it completely impossible to refer to Python dunder methods. Please fix. Thank you. Best --- ## [Introduce Volume-Based Discounts for Codex Credit Purchases](https://community.openai.com/t/introduce-volume-based-discounts-for-codex-credit-purchases/1407356)
**Author:** [@chancex6](https://community.openai.com/u/chancex6)\ **Replies:** 0\ **Last updated:** [October 17, 2026, 11:57pm UTC](https://community.openai.com/t/introduce-volume-based-discounts-for-codex-credit-purchases/1407356 "2026-10-17T23:57:40Z")
THE CORE PROBLEM I would like to propose a volume-based pricing model for purchasing additional Codex credits. The central issue is simple: when larger credit packages maintain the same effective price per credit as sm… --- ## [Codex Pro 5x weekly limit exhausted in 2 days with unchanged workflow + purchased credits not applied](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726)
**Author:** [@kharitonova.pro](https://community.openai.com/u/kharitonova.pro)\ **Replies:** 6\ **Last updated:** [October 17, 2026, 10:04pm UTC](https://community.openai.com/t/codex-pro-5x-weekly-limit-exhausted-in-2-days-with-unchanged-workflow-purchased-credits-not-applied/1390726 "2026-10-17T22:04:06Z")
I would like to report a significant change in Codex usage behavior. I have been using Codex continuously since April/May 2026. My workflow and project complexity have been consistent: I use Codex for developing my appl… --- ## [Building on a student budget when AI tools tell you to take a nap](https://community.openai.com/t/building-on-a-student-budget-when-ai-tools-tell-you-to-take-a-nap/1407302)
**Author:** [@BartMhan](https://community.openai.com/u/BartMhan)\ **Replies:** 0\ **Last updated:** [October 17, 2026, 7:34pm UTC](https://community.openai.com/t/building-on-a-student-budget-when-ai-tools-tell-you-to-take-a-nap/1407302 "2026-10-17T19:34:07Z")
Hey everyone! :waving\_hand::ghana: I’m a student developer from Ghana, currently building a software product, and man, sometimes this journey is genuinely hilarious. :joy: Building something ambitious while being a stu… --- ## [Feature Request: Allow normal ChatGPT conversations to be dragged or moved into Codex-originated/code projects](https://community.openai.com/t/feature-request-allow-normal-chatgpt-conversations-to-be-dragged-or-moved-into-codex-originated-code-projects/1397101)
**Author:** [@BraveFjord99](https://community.openai.com/u/BraveFjord99)\ **Replies:** 2\ **Last updated:** [October 17, 2026, 1:17pm UTC](https://community.openai.com/t/feature-request-allow-normal-chatgpt-conversations-to-be-dragged-or-moved-into-codex-originated-code-projects/1397101 "2026-10-17T13:17:05Z")
I’d like normal ChatGPT conversations to be addable to code projects originally created from Codex. At the moment, there seems to be an inconsistency in how this works: Conversations created in Work mode appear to be… --- ## [28 Days of Shipping: Can we finally get project-scoped chat history in the Codex VS Code extension?](https://community.openai.com/t/28-days-of-shipping-can-we-finally-get-project-scoped-chat-history-in-the-codex-vs-code-extension/1407198)
**Author:** [@fizgen](https://community.openai.com/u/fizgen)\ **Replies:** 0\ **Last updated:** [October 17, 2026, 11:42am UTC](https://community.openai.com/t/28-days-of-shipping-can-we-finally-get-project-scoped-chat-history-in-the-codex-vs-code-extension/1407198 "2026-10-17T11:42:04Z")
One of the most frustrating parts of using Codex in VS Code is how conversation history is organized when working across multiple projects. My workflow is straightforward: Open WSL. Navigate to a project directory… --- ## [When do you cut your losses and start a fresh chat?](https://community.openai.com/t/when-do-you-cut-your-losses-and-start-a-fresh-chat/1407135)
**Author:** [@auscaster](https://community.openai.com/u/auscaster)\ **Replies:** 5\ **Last updated:** [October 17, 2026, 9:29am UTC](https://community.openai.com/t/when-do-you-cut-your-losses-and-start-a-fresh-chat/1407135 "2026-10-17T09:29:02Z")
I’ve had Codex explain exactly why it went off track, agree on the correction, then repeat the same mistake a few messages later. I keep steering because it already has the context. But that context includes whatever as… --- ## [Codex CLI TUI fails during account/read workspace requirements bootstrap (-32603)](https://community.openai.com/t/codex-cli-tui-fails-during-account-read-workspace-requirements-bootstrap-32603/1407111)
**Author:** [@abPlaid](https://community.openai.com/u/abPlaid)\ **Replies:** 0\ **Last updated:** [October 17, 2026, 4:32am UTC](https://community.openai.com/t/codex-cli-tui-fails-during-account-read-workspace-requirements-bootstrap-32603/1407111 "2026-10-17T04:32:44Z")
Environment Surface: CLI Codex version: codex-cli 0.163.0 OS and IDE/browser, if relevant: macOS Terminal, Node.js v22.23.1 through NVM Model and plan or API-key authentication: gpt-6.1-sol, ChatGPT Plus authentication … --- ## [No Text-to-Speech Option Available in Codex Mobile](https://community.openai.com/t/no-text-to-speech-option-available-in-codex-mobile/1402661)
**Author:** [@james-kerfoot](https://community.openai.com/u/james-kerfoot)\ **Replies:** 2\ **Last updated:** [October 16, 2026, 10:18pm UTC](https://community.openai.com/t/no-text-to-speech-option-available-in-codex-mobile/1402661 "2026-10-16T22:18:24Z")
Apparently this has been a limitation for almost 5 months. This is a non-negotiable feature for me. Claude’s text-to-speech is weird, but I am about to cancel my ChatGPT subscription due to this limitation and move to Cl… --- ## [Orchestrator agent - visual code extension](https://community.openai.com/t/orchestrator-agent-visual-code-extension/1402349)
**Author:** [@tomtom0013](https://community.openai.com/u/tomtom0013)\ **Replies:** 2\ **Last updated:** [October 16, 2026, 9:25pm UTC](https://community.openai.com/t/orchestrator-agent-visual-code-extension/1402349 "2026-10-16T21:25:31Z")
We’re building Agent Watch, an open-source VS Code extension that warns when two Codex sessions work on the same file. We want users to keep using the official Codex extension. Is there a supported, read-only API that l… --- ## [Built-in Browser history is recorded internally but not visible or suggested](https://community.openai.com/t/built-in-browser-history-is-recorded-internally-but-not-visible-or-suggested/1390677)
**Author:** [@Alcatraz69](https://community.openai.com/u/Alcatraz69)\ **Replies:** 1\ **Last updated:** [October 16, 2026, 7:58pm UTC](https://community.openai.com/t/built-in-browser-history-is-recorded-internally-but-not-visible-or-suggested/1390677 "2026-10-16T19:58:03Z")
Hi everyone, I’m experiencing a persistent issue with the built-in Browser in the ChatGPT Windows desktop app, version 26.810.7004.0. The Browser is correctly recording visited pages internally: the session history can… [Next page](https://community.openai.com/tag/codex/6.md?match_all_tags=true&page=1&tags%5B%5D=codex)