Codex 28 · Day record · 10 Oct 2026

Day 6 of 28

Closed at 00:00 UTC on 11 Oct 2026 Counted only from activity timestamped inside this UTC day. Later votes, answers and experiences do not change it; only an entry in the corrections log can.

Revised 25 Oct 2026, 06:19 UTC after an entry in the corrections log. Corrections log →

Download card

The day’s change

Recorded update

Improvement

Day 6: Dots can start and follow up on Codex work and edit Scheduled Tasks

Tibo says dots got a pretty big upgrade, quoting ChatGPT's announcement that a dot can now start work in Codex and follow up on existing threads, drawing on ChatGPT conversations, Codex threads and automations. The announcement also says dots are better at deciding whether to continue a thread or start a new one, and can review and edit Scheduled Tasks in ChatGPT Work.

Read the whole note

Tibo's post was published on 10 October at 04:16 UTC and names no campaign day, so this entry follows the UTC date as Day 6. The quoted ChatGPT announcement was published on 9 October. These delegation and task-management capabilities are distinct from the mobile dot creation recorded on Day 5.

This is a feature announcement, not a usage reset or a benchmark. It does not establish that reported dot connection or active-thread attachment errors are fixed. Check by asking a dot to start or follow up on Codex work, or to review and change a Scheduled Task. Rantr has not independently verified availability, reliability or the claimed improvement in thread selection.

Official ChatGPT announcement ↗

Source ↗

Fewer than 10 votes by the close of the day; the tally is in the open data, not shown here.

Improvement

Day 6.2: Connect ChatGPT Go, Plus or Pro to Devin, with GPT usage drawn from the plan quota

Tibo says a ChatGPT subscription is now also a Devin subscription, quoting Cognition's announcement that personal ChatGPT Go, Plus and Pro plans can be connected to Devin. Cognition says all GPT model usage in Devin draws down from the connected plan's quota.

Read the whole note

Tibo's post was published on 10 October at 04:33 UTC and names no campaign day, so this is recorded as an additional Day 6 item. The quoted Cognition announcement was published on 9 October. This adds a way to use a personal ChatGPT plan in Devin; it is not a usage reset, a separate quota allowance or a benchmark, and the post does not describe coverage for non-GPT models.

Check the ChatGPT plan connection in Devin and the resulting quota usage. Rantr has not independently verified plan availability, connection behavior or quota accounting.

Official Cognition announcement ↗

Source ↗

Fewer than 10 votes by the close of the day; the tally is in the open data, not shown here.

Reset

Day 6: Tibo announces a reset as the Day 6 delivery

Tibo's post reads, in full: "Day 6/ Reset". It was published on 10 October at 22:15 UTC, the last hours of Day 6, and names the day itself, so it is recorded as Day 6's reset alongside the Day 6 Dots improvements.

Read the whole note

The post gives no further detail: it does not say whether the reset is applied immediately or banked for later use like the Day 3 reset, which plans receive it, or when it expires. This is a usage reset as Tibo announces it, not a change to the limits themselves.

Check your usage panel for a refreshed balance or a new banked reset and its expiry, then share what changed. Rantr has not independently verified the reset or its application to individual accounts.

Follow-up from Tibo on 10 Oct (x.com ↗): "Global reset by EOD. May the tokens do good things for you." Scope: the reset is global and applied by the end of the day, not banked for later; he does not name a time zone, the plans covered or an expiry. Rantr has not independently verified.

Companion post from Tibo on 11 Oct (x.com ↗): "Put the reset in the bag. All propagated." Posted at 04:01 UTC quoting his "Global reset by EOD" follow-up, it closes the Day 6 reset, so it stays on Day 6 rather than counting for Day 7. Scope: he says the reset has finished reaching accounts; he still names no plans or expiry. Rantr has not independently verified that every account received it.

Source ↗

Fewer than 10 votes by the close of the day; the tally is in the open data, not shown here.

Running totals

What changed since day 5

Totals since 5 Oct 2026, at the close of each day; today’s are so far. A total never goes down when a sample is recounted.

Replies on X

Replies to Tibo’s post published this day

A sample collected from X, not every reply. Each link opens the reply on X.

Replies dated this day in the collected sample: 191. Collected since 5 Oct 2026: 861.

X replies are found by search and each one is checked against its public post on X. Last read: 26 Oct 2026, 23:46 UTC. Not every reply is read.

4 X replies from this day are not quoted: Rantr does not have their words or held them back after automated screening. They stay on X.

Corrections log

Corrections logged this day

  1. · Update

    Day 5.2: Create and text a dot entirely in the ChatGPT mobile app

    Before
    The closed Day 5 record listed composer predictions but omitted Tibo’s separate mobile dot creation and texting announcement.
    After
    Day 5 now also records mobile dot creation and texting as Day 5.2, alongside composer predictions; no new day or usage reset is added.
    Why
    Late recording of Tibo’s Day 5 dots announcement, posted on 9 October at 19:49 UTC: https://x.com/thsottiaux/status/2108646052178092403. The live post announces a different feature from composer predictions, rather than a formatting revision; the closed day record needs to include it.

  2. · Update

    Day 4: Tibo announces instant steering in Codex and releases GPT-6.1 Sol ultrafast

    Before
    These are two improvements, faster steering mid-run and a faster model option; neither is a usage reset, and the 8x figure is OpenAI's own claim, not a benchmark. Check it by steering a running Codex task and choosing GPT-6.1 Sol ultrafast in Codex, ChatGPT Work or the API as the rollout reaches you. Rantr has not independently verified the speed or the steering change.
    After
    On 9 October, replying to a report of slow GPT-6.1 Sol ultrafast output in Playground, Tibo says the issue was “all client-side + network for playground only” and that optimized rendering logic has shipped to keep up with ultrafast. This follow-up concerns Playground rendering. It does not establish that Codex speed complaints are fixed. Rantr has not independently verified the rendering change or its effect on speed. Source: https://x.com/thsottiaux/status/2108428560822424062
    Why
    Added Tibo’s Playground-only rendering follow-up to the existing Day 4 announcement. It is not a new release or Day 5 announcement, does not prove Codex speed complaints are fixed, and has not been independently verified by Rantr.

  3. · Update

    Day 3: Tibo loads a banked reset in every paid account

    Before
    Tibo's confirmation: https://x.com/thsottiaux/status/2108040921044639779
    After
    In a reply to the banked-reset announcement, Tibo said “Will be there by EOD PST.” This was an arrival estimate, not an additional reset. His later confirmation that it landed across all accounts is already cited above. Source: https://x.com/thsottiaux/status/2107913738791596286
    Why
    Added the earlier banked-reset arrival estimate alongside the already-recorded landing confirmation. This closes the missed follow-up without counting another reset; individual accounts remain unverified.

  4. · Update

    Day 2.4: Decisions API enters public beta

    Before
    Tibo announced the Decisions API as Day 2.4, quoting OpenAI Developers: developers can use it to choose a model, tool, or action in near real time. OpenAI describes it as available to all developers in public beta and claims decisions up to 10x faster than GPT-6 Luna through the Responses API; this is a vendor comparison, not an independent benchmark. Tibo says they will use it to improve the app experience. That app integration is a future intention, not a confirmed shipped app feature or usage reset.
    After
    Replying to a question about whether the Decisions API accepts images, Tibo answers “It does”. This confirms his statement about image input for that API; it does not announce a new Codex model or another daily release. Rantr has not independently tested image input. Source: https://x.com/thsottiaux/status/2107628890550399465
    Why
    Added Tibo’s confirmation that the Decisions API accepts images, with the question and reply as context. Rantr has not independently tested it; the existing Day 2.4 release is unchanged.

  5. · Update

    Day 2.1: Auto-review is now free with ChatGPT sign-in

    Before
    Tibo later clarifies that Auto-review is the “Approve for me” permission mode: open the permissions menu below the composer and select “Approve for me”. This adds his later UI path to the settings path in the announcement; Rantr has not independently checked it in every app version. Source: https://x.com/thsottiaux/status/2107504197012988007
    After
    In the Day 2 roundup, Tibo calls Auto-review “Approve for me”, says it is included without using plan allowance, and says it could previously account for 2–10% of a plan when used. That range is Tibo’s claim, not a measurement by Rantr. Source: https://x.com/thsottiaux/status/2107575657014468879
    Why
    Added the Day 2 roundup’s Auto-review allowance range with attribution. The other roundup items are already recorded separately, so this adds context to Auto-review without another release.

  6. · Update

    Day 2.1: Auto-review is now free with ChatGPT sign-in

    Before
    In a follow-up to the Auto-review announcement, Tibo recommends using Auto-review instead of full access and calls it “not a trade-off anymore”. This is his recommendation, not Rantr’s independent security assessment. Source: https://x.com/thsottiaux/status/2107370852564021582
    After
    Tibo later clarifies that Auto-review is the “Approve for me” permission mode: open the permissions menu below the composer and select “Approve for me”. This adds his later UI path to the settings path in the announcement; Rantr has not independently checked it in every app version. Source: https://x.com/thsottiaux/status/2107504197012988007
    Why
    Added Tibo’s later clarification of the “Approve for me” name and composer permission-menu path. This revises the existing Auto-review entry; no new release is counted.

  7. · Update

    Day 2.1: Auto-review is now free with ChatGPT sign-in

    Before
    OpenAI’s technical explanation: https://alignment.openai.com/auto-review/ Try it on your own tasks and vote on whether this change is useful.
    After
    In a follow-up to the Auto-review announcement, Tibo recommends using Auto-review instead of full access and calls it “not a trade-off anymore”. This is his recommendation, not Rantr’s independent security assessment. Source: https://x.com/thsottiaux/status/2107370852564021582
    Why
    Added Tibo’s follow-up recommendation to use Auto-review instead of full access, with attribution. Rantr has not independently assessed its security.

  8. · Update

    Tibo announces roughly 50% faster default speeds

    Before
    In a reply to the Day 1 announcement, Tibo adds that speeds are reaching 50 tokens per second instead of 30 and credits the tokenizer and token efficiency. These are Tibo’s figures, not a Rantr benchmark. Source: https://x.com/thsottiaux/status/2107159119107146237
    After
    In a later reply about Astra performance, Tibo says the Astra changes were rolled out earlier and that 50 tokens per second is the expected figure. This clarifies the rollout described above; it is not a separate release or proof of the speed in every account. Rantr has not independently verified it. Source: https://x.com/thsottiaux/status/2107295089416708515
    Why
    Added Tibo’s clarification that Astra changes had already rolled out. The Day 1 entry remains an attributed announcement, not a verified benchmark or a new release.

  9. · Update

    Tibo announces roughly 50% faster default speeds

    Before
    Announced on 5 October at 17:20 UTC, with no user action required and rollout expected within two hours (by approximately 19:20 UTC). This is the announced performance improvement; Rantr has not independently benchmarked it. Try it on your own tasks before judging the result.
    After
    In a reply to the Day 1 announcement, Tibo adds that speeds are reaching 50 tokens per second instead of 30 and credits the tokenizer and token efficiency. These are Tibo’s figures, not a Rantr benchmark. Source: https://x.com/thsottiaux/status/2107159119107146237
    Why
    Added Tibo’s Day 1 throughput figures with their source. They are attributed to Tibo; Rantr has not independently benchmarked them.

Nothing this day for: OpenAI release notes, published experiences.

Method

How this record is made

Each day runs from 00:00 to 24:00 UTC. The record counts votes, release check answers, experiences, “Same here” and replies timestamped inside the day. When the day closes the record is frozen; it is recomputed only after an entry in the public corrections log, or when the day’s update is recorded late. Rantr is a neutral record: it does not take a side.

Corrections log → Open data and method →