Reset forecast

Follow Tibo’s public posts, track Codex usage-limit reset forecasts, and compare predictions with manually verified outcomes.

Published Sep 23, 2026, 00:01 · UTC+8

Experimental forecast; not yet calibrated against historical data.

Prediction accuracy

—

No eligible reviewed time forecasts yet

Based on whether resets fall within the predicted time range

How this is measured

Only scheduled forecasts with a time range saved at publication count, once each. A reset within that range is a hit; an earlier reset or no reset by the deadline is a miss. Unreviewed, uncertain and retrospectively inferred ranges are excluded. Excluded historical records without a previously issued time range: 1. Samples may overlap; small-sample accuracy does not predict future performance.

Next reset · Estimated time

Not enough information yet

No valid time forecast has been published. Waiting for new signals.

Tibo’s posts on X 23

@thsottiaux · Newest first · Original text

T
Tibo@thsottiaux
Reply context

Are you happy with your Codex

👁️codex👁️

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

the GPT-6 Community Night was 🔥

3am on a tuesday

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

I used Codex to vibe code a game on my living room wall! I control it with my hand using computer vision + projection mapping 😁 I'll post a timelapse + my process in the replies!

That's cool!!!

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

i don't get the tibo hate. he resets codex usage regularly whenever there are actual issues, sometimes just for fun. and the team is pretty transparent about the limits. asking for resets as a joke is one thing. swearing at him under every post and acting like he personally stole your limits is another.

Most folks have been very nice! Love this community

Hint signal 10%View original
T
Tibo@thsottiaux
Reply context

astra has been pretty solid for me after i took my time with the agents.md it's not perfect, nothing is, but i like it a bit more than fable 5.1 right now controversial i know

Mind sharing your instructions?

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

OpenAI could save billions of dollars if they could make Astra stop writing the most useless tests imaginable

Billions

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

as the prophecy foretold, we must now seize our golden opportunity

You missed the long way to go part of the quote

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

I thought Tibo commented on a post of mine But then I took a closer look and noticed something off He’s jacked - what the heck 😂

Clearly not me

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

I made a game where you organize an entire survival bunker. Thousands of supplies dumped on the floor, and you shelve them one at a time. It's done when it's clean. Game built with AI using GPT-6 Astra on @TesanaAI

People will do anything instead of cleaning their house

Hint signal 0%View original
T
Tibo@thsottiaux

2026 is the year of linux desktop

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

met a dude, he still copy-pasting the code from chatgpt

did you bail or at least show him codex before?

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

Did Astra suddenly get way more efficient, or is it just me?

Always improving

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

the "astra stopping too early" problem is completely solved by jev btw

I could see Astra and Jev being best friends in real life

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

ok tibo you guys didn’t ship anything interesting this week you owe us a banked reset sorry i don’t make the rules

OK fine. But it’s also still coming in Tuesday

Hint signal 70%View original
T
Tibo@thsottiaux
Reply context

If you want to see an example on how having the very best AI model and harness doesn’t result in better/faster execution - in fact it slows you; It took the Claude Code team 16 months LONGER (!!) to add support for AGENTS.md than OpenCode & Codex, 12 more vs Cursor

Let’s celebrate this change! Positive reinforcement!

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

thanks for reposting Tibo

Tibo from wish dot com

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

i'm a little bit disappointed with OpenAI. I'm running out of my weekly usage limit: • I can't upgrade to the $200 plan • I can't use Astra, and I haven't used it all week • Despite all of this, there has been no reset this week In less than 1h, I'll be forced to switch to Claude again. All of this in my first week as a $100 user. Not good at all

You main Sol? And out of curiosity, you never buy credits?

Hint signal 10%View original
T
Tibo@thsottiaux

We were working on the keynote today with @romainhuet and @sama and most of the fun was trying to figure out how to explain it all to you because there is so much good stuff in there that it's a bit ridiculous all in quick succession. We'll have some things next week already to not keep you waiting so long, but very excited to show you all new things we've been working on and how it will all come together in the coming months.

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

I want GPT-6 SOL AND GPT-6 LUNA

What else do you want

Hint signal 0%View original
T
Tibo@thsottiaux
Reply context

instinct is incredibly annoying I swear. Like the noise to signal ratio is insane

I thought you were the instinct noah and was confused

Hint signal 0%View original
Grok · X collection: FailedGrok · Signal analysis: NormalRefreshes every 30 secondsLast successful collection: Sep 22, 2026, 07:53