capocoincoin

follow to taste the secret sauce

[capocoincoin @PoleSoude]

One small step for me, a giant step for local audio.
Introducing Oratioviva the app to manage and use tts and clone models. The audio of the trailer was made with it. (Its a bit weird bcs I had to speed it up) https://t.co/QlajRBTjuQ

[capocoincoin @PoleSoude]

Hey @thsottiaux @ChatGPTapp Codex systematically crashes whenever the integrated browser is used. The in-app browser/WebView cleanup appears to bring down the entire app. Can I DM someone the session ID and exact timestamp so the Codex team can investigate?

[capocoincoin @PoleSoude]

thx @elonmusk ig

[capocoincoin @PoleSoude]

@JaydenDavisNC Yep

[capocoincoin @PoleSoude]

@LexnLin How many tokens were used?

[capocoincoin @PoleSoude]

Hey @thsottiaux I have more than two brain cells so my usage limit is fine. Can I have the option to keep the old thinking budget?

[Lentils @Lentils80]

🚨 GPT-5.6 Sol's juice values (thinking budgets) have been severely degraded compared to release day

If Sol now feels faster and more "efficient", this is probably why

Terra and Luna juice values aren't affected, so their thinking budgets are now higher than Sol's

[capocoincoin @PoleSoude]

@Laz4rz @sama You need to install the new ChatGPT app on your device, it did it to me before I installed the app and no, it’s fixed

[capocoincoin @PoleSoude]

I can’t imagine how inefficient you have to be to have this happen i’ve got three tasks running on xhigh and this doesn’t happen to me (most of them are long goals) so I guess rewrite your agent MD

[Chubby♨️ @kimmonismus]

Seriously, this has to stop. I've now set GPT-5.6 from high to medium (not fast mode, of course), and I'm still burning through my rates at an insane rate. My 5 hours are almost gone -again.

I've already used up all three resets. OpenAI needs to work on its efficiency. This is the biggest bottleneck.

[capocoincoin @PoleSoude]

I see people complaining about rate limits on the Pro 20x like this

[capocoincoin @PoleSoude]

@chrisgpt There’s a new ChatGPT beta app found it by accident on Windows store trying to update codex

[capocoincoin @PoleSoude]

Rare @FrenchResponse L meae is biased ofc

[French Response @FrenchResponse]

They skipped the "illegal content" part.
It makes the story less dramatic.

[capocoincoin @PoleSoude]

@DFintelligence tu penses sry

[capocoincoin @PoleSoude]

@DFintelligence Base sur ce que t'as vu en interne Je pense qu'on a encore combien l'énumération avec les Européens soient complètement coupés Il faudrait que j'aille aux us ou autre Mais je sais pas comment

[capocoincoin @PoleSoude]

@chetaslua isnt 5.6 sol and pro sol ultra?

[capocoincoin @PoleSoude]

@AndrewCurran_ And what would we need to do in order to avoid such scenario?

[capocoincoin @PoleSoude]

@AndrewCurran_ Do you think we’re eventually gonna see the appearance of an extremely powerful and authoritarian regime in the US because the next logical step is to progressively diminish the access to the best models even to the highest members of society in order to completely control them

[capocoincoin @PoleSoude]

@mirochill ok top dhab chetas est plus rapide je viens de voir que tt chez dev mode vous faites du bon boulot. des tips pour avoir contacts/access?

[capocoincoin @PoleSoude]

@chetaslua Did you find it token efficient aka is it ever gonna be somewhat useful for regular ppl 80 is ok but still

[capocoincoin @PoleSoude]

Stop spreading bs they r separated oai would never even if they did big acc would’ve talked about it already @chetaslua @chrisgpt etc

[Jahir Sheikh @jahirsheikh8]

First Anthropic linked everything to one usage limit.

Now OpenAI is doing the same 😭

My ChatGPT usage is impacting my Codex daily quota already 🤧

[capocoincoin @PoleSoude]

Only sub that won’t lose value @scaling01

[Tibo @thsottiaux]

@tmuxvim We actually have amazingly efficient models