Yazan Al Khatib

@Entrepenulian @theo Ikr! Now all we need is 2 TB of RAM to run it locally lol

@yzkhatib @theo I love how the Chinese just let you do what ever you want with their models.

What are the best harnesses to use with Kimi K3 right now? It heard it's pretty good at orchestration, where's the best place to see that?

@Entrepenulian @theo You can Kimi has an Anthropic style endpoint

@theo Can we not use K3 in Claude Code somehow? Like Theo did with Sol in CC?

@ajambrosino @thsottiaux It's Sol good man

@thsottiaux better col Sol

How do you pronounce Sol

@isiliconx @xxwang99999 @thsottiaux Yeah, exactly! Because business plans are certainly more expansive than Plus and can even scale up to be more expensive than Pro

@yzkhatib @xxwang99999 @thsottiaux We do even have access to 5.6/5.5 Pro.

Another reset for our Codex and ChatGPT Work users. Actually hit 9M active users way earlier today, but then got distracted by the approximately millions of things the team is doing to keep the systems up and reliable. Should have that sweet 100% weekly usage limit back in a few minutes. Go be your productive self and close twitter. Shoo!

@theo Terra is usable imo ( for certain tasks )

My recommendation: just use Sol.

I'm considering purchasing a new account for either ChatGPT or Claude 20x, what do you suggest I get? I believe Fable is smarter than 5.6 Sol, but since Fable tends to flag most basic tasks and then hands them over to Opus 4.8 ( opus currently comparable to 5.6 Terra )I’m second guessing whether I should go for another claude account

@hey_yogini I do agree with this proposal

please start sharing these plans in advance so I can plan my work accordingly

@YashHustle_22 Fable 5 is smarter and faster, 5.6 Sol has a slightly better taste

Which AI model do you prefer right now? - GPT-5.6 - Fable 5

@oneuios Nice

Liquid glass on Instagram too. What did you think? I loved it😍

@buildwithsid That's what competition does

this cant be a coincidence... im not complaining lol keep going

@hqmank Me too

GPT-5.6-sol in Codex has been painfully slow today. 3+ minutes per response. Thought it was my network. Cursor and Claude Code both respond fine. Anyone else getting this?

@thsottiaux @rezoundous Bro he means ChatGPT, now can you give our business accounts a reset? They're more expensive than plus

@rezoundous OpenAI?

Has OpenAI reached 9m users yet? I need a reset lol

@xxwang99999 @thsottiaux Me too, although we pay more than Plus users per seat, so we should be treated as pros (not including extra payments for purchasing credits).

@thsottiaux Why wasn't ChatGPT Business reset?

@ziwenxu_ Yup exactly!

@yzkhatib As long as it finishes the task haha

Our /goal in Codex has been running for over 20 hours.

@Dayanch96 @OpenAI @OpenAIDevs Also it may be “for our employees only“, Peak marketing

@yzkhatib @OpenAI @OpenAIDevs it is for all selected users xD

@OpenAI @OpenAIDevs Hi there OpenAI, my account did not get the usage reset yesterday, is this intentional? Because I contacted support and they said it is for selected users, but has been announced that it is for all users

@heliumbrowser How long are you planning on maintaining the support ?

helium still supports mv2 extensions btw

@elonmusk wow

Once we have completed our review for security vulnerabilities, we will make the entire codebase of 𝕏 open source, with no exceptions. Moreover, we will invite third party reviewers to examine the system that is running to confirm that the open source code is what is running. Trust through total transparency is the only thing that should be believed.

@thsottiaux Mine did not reset and it's showing now my weekly limit reset on the 20th it was on the 19th

Embarrassment of riches. But looks like we might hit 9M soon. Should we reset the ChatGPT Work and Codex usage again or give it some space?