Codex Update

Banked reset 2 of 2 is live. Day 28, the final day of Tibo's 28-day run. Like the first one, it doesn't refill anything until you redeem it: - Desktop app: Settings > Usage & billing - Web: Usage settings - CLI: /usage, then Redeem reset - Not in the mobile app Plus and Business get the 5-hour and weekly allowances back, Pro gets the weekly. It only redeems once an allowance is under 80% remaining, and it expires, so check the date shown in Usage. One gotcha: with "Allow your assistant to use resets" on in the desktop app, Codex can spend it for you at 10% or less. Turn it off if you want to pick the moment. help.openai.com/en/articles/…

3 banked-reset gotchas to know before you redeem yours: 1. Codex can spend it for you. With "Allow your assistant to use resets" on in the desktop app, it can redeem one once your 5-hour or weekly allowance is at 10% or less. Turn it off if you want to choose the moment. 2. Resets don't stack. If a global reset lands after you redeem, you go back to 100%, not 200%, and the banked one isn't returned. 3. Plus and Business: check both bars first. The 5-hour and weekly periods end at different times, and a reset only restores allowances below 100%. To see when one was used: desktop app, Settings > Usage & billing > History (last 30 days). help.openai.com/en/articles/…

9 stable releases and a 28-day run: Codex's October, all in one post. Everything that shipped or changed in October 2026, checked against GitHub, OpenAI's changelog, help pages and official posts. Bookmark it for reference. 1. Biggest launches - GPT-6.1 Terra came to Codex on Oct 26 for Plus, Pro, Business, Enterprise and Edu. Tibo says it gives 2x the usage of 6.1 Sol. - GPT-6.1 Sol Ultrafast arrived Oct 8. OpenAI says it's up to 8x faster than Sol Standard. Pro $500 and eligible Enterprise and Edu only. - Sol's default context in Codex went from 272k to 500k tokens on Oct 11, with no long-context fees on a ChatGPT plan. - Tibo started a 28-day run on Oct 4: one clear improvement or a full limits reset every day. - Also shipped: composer predictions (beta, Pro), a re-shipped Codex cloud with Tailscale support, a new Windows sandbox built on Microsoft's MXC, editing earlier messages in the desktop app, and about 60% less app data on disk plus a Storage tab. 2. Every stable CLI release - 0.160.0 (Oct 1): start sessions outside a project, fewer SQLite stalls, Windows sandbox fixes - 0.160.1 (Oct 5): keeps Windows env vars for remote MCP servers - 0.161.0 (Oct 7): GPT-6.1 Sol becomes the default model, /mcp login, mic and speaker picker - 0.162.0 (Oct 8): managed Git worktrees, pin tasks with p, /copy, clickable links - 0.162.1 (Oct 9): fixes a TUI crash and background-server startup failures - 0.163.0 (Oct 15): searchable /config, steer mid-response, leader key in /keymap, /daemon restart - 0.164.0 (Oct 23): ten /presets slots, inline /model and /permissions, image previews - 0.165.0 (Oct 27): built-in Vertex AI provider, Terra in the catalog, memory v2 on by default - 0.166.0 (Oct 29): codex doctor --network-only, transcript search, codex queue --resume 3. Pricing, plans and limits - Full limit resets landed again and again during the run, including Days 6, 8, 10, 13, 15, 21 and 23. The run closes with banked resets you redeem yourself. - Auto-review ("Approve for me") stopped counting against usage (Day 2). Day 12 stretched usage with longer caching, and Day 16 made the cache keep with no expiry for logged-in users. - GPT-5.5 left Codex for ChatGPT sign-in on Oct 14. API-key sessions kept it. - About 40,000 accounts had limits that didn't match their plan. OpenAI fixed it and added credits. Pro 500 upgraders hit by a separate bug got 25,000 credits. - Codex for OSS grants moved up to the Pro 200 tier. - Oct 29: limits doubled for Pro 100, 200, 500 and Business Premium, and the planned Pro 200 cut was cancelled. - Pro and Business now get monthly API credits. They don't add to your Codex allowance. - New device limits cap how many paid accounts one machine can hold ("Device limit reached"). - Coming in December: ChatGPT and Codex share one token-based allowance on Plus, Pro and Business. 4. What people hit, and the fixes - Sudden usage drain (mid-month): a confirmed bug, now fixed. If it happens again, run /feedback in the affected session and check /usage daily. - "Selected model is at capacity" or 429s with quota left (GitHub #57084): no official fix yet. Switch with /model, then codex resume to continue. - VS Code extension 26.1021 won't start its UI: users report the pre-release build, rolling back with Install Specific Version, or launching with code --disable-gpu. - Worktree chats fail with "Cannot serialize value" on Windows and Linux: Local mode still works. - Max or Ultra missing from the reasoning slider (#58201): re-check it in Settings > General > Available reasoning efforts. - Usage rings gone from the desktop sidebar (#57656): use Settings > Usage or /status. - Security: don't launch Codex inside ~/Downloads, where Forever Security found a planted AGENTS.md. Don't install from reseller one-liners either. Symantec caught one turning the sandbox off. 5. Three tips worth keeping - Get a baseline before December's shared pool: /usage daily and /usage weekly in the CLI. - Pipe anything into Codex: npm test 2>&1 | codex exec "summarize the failures and propose the smallest fix" - Mind the AGENTS.md cap. Codex merges your AGENTS.md files and stops at 32 KiB (project_doc_max_bytes), so split rules into nested folders. Ask for November: publish the per-plan token allowance before the shared pool starts, so users can do the maths. Sources: github.com/openai/codex releases and issues, developers.openai.com/codex/…, help.openai.com, and posts from @OpenAI, @OpenAIDevs and @thsottiaux

Good question. OpenAI's help page doesn't give an exact number, so we can't say what the cap is. It only says one device can use so many paid personal accounts (Go, Plus, Pro), and Business, Enterprise, Edu and Free don't count.

Fair question. OpenAI's help page says device limits are separate from account switching limits, so switching still works. The cap is on how many paid personal accounts (Go, Plus, Pro) one device can use. The page is linked at the bottom of the post.

Codex this week: 9 changes, 5 user pain points, all here. Oct 25 to Nov 1, checked against OpenAI's posts, help pages and GitHub. Releases 1. GPT-6.1 Terra rolled out in Codex for Plus, Pro, Enterprise and Edu (Oct 26). Tibo says it gives 2x the usage of 6.1 Sol. 2. CLI v0.165.0 (Oct 27): built-in Vertex AI provider, change reasoning effort mid-chat, memory v2 on by default. 3. CLI v0.166.0 (Oct 29): Terra in the Vertex AI and Bedrock catalogs, codex doctor --network-only, transcript search, and a fix for early compaction after resume or fork. 4. Text your dot over iMessage or RCS (Pro, US preview). Multi-account switching landed on mobile, with desktop "coming soon". Limits and pricing 5. Limits doubled for Pro 100, 200, 500 and Business Premium, and the Oct 30 Pro 200 cut was cancelled. 6. Pro and Business now get monthly API credits ($100 to $500 on Pro). They don't raise your Codex allowance. 7. Confirmed: from December, ChatGPT and Codex share one token-based allowance on Plus, Pro and Business. 8. New device limits: one machine can only hold so many paid accounts, or you get "Device limit reached". 9. Banked reset 1 of 2 is live. Redeem it in Usage settings or with /usage in the CLI. Incidents and security - Oct 27: GPT-6.1 Sol was degraded for about 78 minutes, then resolved. - Don't run Codex inside ~/Downloads (Forever Security found a planted AGENTS.md), and don't install it from reseller one-liners (Symantec caught one turning the sandbox off). - Forever Security says a local HTML page opened in Codex's built-in browser can read files on disk. Still unpatched as of Oct 29, per their write-up. What users are saying - 429 and "at capacity" errors with most of the allowance left, mostly VS Code on Linux. The busiest new GitHub issue this week (#57084, 87 comments). No fix yet. Switch models with /model, and codex resume picks the session back up. - VS Code extension 26.1021.22027 "could not start its user interface" on Linux and Windows, across several issues. Reported workarounds: the pre-release build, an older version, or launching with code --disable-gpu. - Usage rings vanished from the desktop app sidebar after recent updates (#57656, Windows and Mac). Check Settings > Usage or /status meanwhile. - Max or Ultra missing from the reasoning slider after 26.1027 (#58201). Fix confirmed by several users: Settings > General > Available reasoning efforts, re-check the level. - The shared December pool is the loudest topic on X, and a GitHub request to keep Plus allowances separate is gaining votes (#58530). What people are asking OpenAI for: publish how chat and Codex tokens will be counted before December, bring the usage rings back, post a status note when 429s hit accounts with quota left, and give a fix date for the built-in browser issue. Community shoutout @talkaboutdesign built agent-stat, which shows your local Codex usage, per-request cost and live limits in one place: npx agent-stat@latest Tip Hold your banked reset until you need it. It only redeems once an allowance is under 80% remaining, and it moves your weekly reset date to the day you use it. Sources: openai/codex releases and issues on GitHub, help.openai.com, @thsottiaux and @OpenAI on X

Banked reset 1 of 2 is live. Tibo just dropped Day 27 of the 28-day run. It won't refill anything by itself. You redeem it: - Desktop app: Settings > Usage & billing - Web: Usage settings - CLI: /usage, then Redeem reset - Not available in the mobile app What it restores: Plus and Business get the 5-hour and weekly allowances back, Pro gets the weekly. Your weekly reset date moves to the day you use it. Save it until you need it. It only redeems once an allowance is under 80% remaining, and it expires, so check the date shown in Usage. How it works: help.openai.com/en/articles/…

Track your Codex token burn before December's shared pool. From December, Plus, Pro and Business share one token-based allowance across ChatGPT and Codex (per OpenAI's help page). Get a baseline now in the Codex CLI: /usage daily /usage weekly /usage cumulative Live limits in your footer: /statusline, then toggle rate limits and context stats Per-session check: /status shows model, context left and token usage /usage is also where you redeem an earned reset if you have one. Bigger models and complex tasks burn more, so a week of numbers tells you if your plan still fits. Docs: developers.openai.com/codex/…

Confirmed: ChatGPT and Codex will share one usage limit from December. OpenAI's help page, updated today: - Plus, Pro and Business get one allowance across ChatGPT and Codex - Usage is measured in tokens, so bigger models and harder tasks burn more - Instant chat doesn't count against it - Combined remaining usage and reset times will show in Settings > Usage - Hit the cap and you wait, upgrade, or buy credits OpenAI says most users won't hit resets more often than now. If you run long Codex sessions and lean on ChatGPT too, plan for both drawing from the same pool. Fair ask to OpenAI: publish the token allowance per plan before December so people can check the maths. help.openai.com/en/articles/…

Right, we're an independent account for Codex users, not OpenAI. Everything in the post comes from OpenAI's own help page, linked at the bottom.

The help page doesn't define "device," so nobody can promise separate Linux users count separately. It does say shared or public devices can hit the limit even when each person uses their own account, and to tell Support if you share one. Business workspace seats don't count toward it.

Existing subs can stay: more on Codex device limits. Lots of you asked follow-ups, so here's what else OpenAI's help page says: - Subscriptions bought before the limits took effect "can continue under the existing-subscription exception" - Cancel one of them and there's no guarantee you can replace it - Blocked at sign-in? The warning offers cancel + refund for ChatGPT and Google Play purchases. Apple buyers request it through Apple - A shared household or public device, or one card used by several people, can trip the limit too. Tell Support if that's you - The exact number of paid accounts per device isn't published Fair ask to OpenAI: publish that number, so people know before they pay. help.openai.com/en/articles/…

Switching is still allowed. OpenAI's page says device limits are separate from account switching limits. The cap is on how many paid personal accounts (Go, Plus, Pro) one device can use. Business, Enterprise, Edu and Free don't count, and the page doesn't give the exact number.

Known bug: Max and Ultra gone from the Codex app slider. After the latest desktop update, the reasoning slider stops at Extra High, even when Settings shows Ultra as your default. 10 people on one GitHub issue so far, on Mac and Windows, with Astra and Sol. Two fixes users confirmed: 1. Settings > General > available reasoning efforts: re-check Max and Ultra. 2. Or add this to ~/.codex/config.toml, then fully restart the app: [desktop] enabled-reasoning-efforts = ["low", "medium", "high", "xhigh", "max", "ultra"] show-ultra-in-model-picker-slider = true Config fix credit: JXN20210043 on GitHub. Fair ask to OpenAI: keep the reasoning levels users picked across updates. github.com/openai/codex/issu…

Heads-up: stacking paid accounts on one machine for extra Codex usage can now get you locked out. OpenAI's new help page on card and device limits: - One device can only use so many paid accounts (Go, Plus, Pro 100/200/500) - Hit it and you see "Device limit reached" at sign-in, or get signed out later, even if the purchase went through - Signing out of another account doesn't free a slot - Business, Enterprise, Edu and Free don't count toward it This lands the same week multi-account switching shipped on mobile, so check before you buy a second sub. The sanctioned route to more usage is credits under Usage & Billing in the Codex app. help.openai.com/en/articles/…

Two more Codex resets are coming: Days 27 and 28 of the 28-day run. Tibo's Day 26 update: - Switching between multiple accounts now works on mobile - Account switching on desktop is "coming soon" - The next two days are both limit resets, and he says they're banked If you've been rationing Codex usage, you can spend it today. The last reset landed around midnight PST, so run /status after that to confirm.

Built with Codex usage in mind: @talkaboutdesign shipped agent-stat 0.3.0. It shows your Codex and Claude Code usage locally, no account: - per-request costs at current model prices - subagent and archived sessions counted - 24h/7d/30d/90d dashboard, cache savings, live subscription limits Useful after this week's limit changes: check what your tasks actually spend. Try: npx agent-stat@latest

Known bug: Codex app dictation got worse in 26.1027.30824. Since the latest app update, users report: - misheard words in Swedish, Russian and English dictation - the live waveform is gone, so you can't tell it's recording - some recordings don't start at all Workaround: one user says rolling back to the previous build fixed it. The same dictation still works in Safari and ChatGPT Classic. @ajambrosino asked for app complaints during the sprint. Fair ask: bring back the waveform and fix non-English accuracy. Also new in Codex: GPT-6 Rosalind Research, for approved life-sciences Research orgs only (gpt-6-rosalind in the API). Source: help.openai.com/en/articles/…

Side chats in the Codex app will be saved soon. App lead Andrew Ambrosino asked if side chats should persist, then answered "vox populi, vox dei. will ship soon." What users asked for in his replies: - Don't lose side chats when the app closes, restarts or updates - A way to turn a side chat into a regular chat, or hand it back to the main one - Keep a temporary mode for throwaway questions Until it ships, treat side chats as disposable and copy anything worth keeping into the main thread before you quit.

Codex tip: diagnose connection drops with one command. Update to CLI v0.166.0, then run: codex doctor --network-only It runs only the config, connectivity and WebSocket checks, so you can see if the problem is on your side (proxy, VPN, config) before restarting sessions. Bonus in the same release: codex queue --resume resumes an unloaded thread after you queue a message. Update: npm i -g @openai/codex@latest Notes: github.com/openai/codex/rele…