Prabh

Repair tech for 15 years | Creator of OmniMend | Artist | Learner 🚀 ☺️
1,000 followers 🙂

[Prabh @arora2992]

@thsottiaux Even on 20x I’m out in a few days….
Barely doing any parallel work….
Astra will be worse?

5.5 was an amazing balance. I could do so many things at the same time and for hours.

[Prabh @arora2992]

I made a software that fixes electronics. Sending 1 month free to businesses. https://t.co/ocj9YMYnEa

[Token Gremlin @TokenGremlin]

You’re seeing a flood of GPT-6 Astra leaks everywhere right now.

Just know that this account right here was the one that originally posted the first leak.

But of course, the algorithm is ruthless, so all the impressions end up going to someone else we both know. Hahaha.

[Prabh @arora2992]

@TokenGremlin Commenting and reposting so that you can get more credit :)

[Prabh @arora2992]

@TokenGremlin Commenting and reposting so that you can get more credit :)

[Prabh @arora2992]

@nickbaumann_ the reset button in sight

[Prabh @arora2992]

@ThePrimeagen nah, definitely needs to be much smarter still

until we can all stop saying, "why d fuck did you do that"

we literally have to prompt engineer the hell out of the current models so that they don't do dumb shit

it's a glorified rube goldberg machine right now

[Prabh @arora2992]

Codex in-built search is dogshit.
I can search for the exact chat thread name and it doesn't show up.

[leo 🐾 @synthwavedd]

🚨 BREAKING: "gpt-6-astra" has been staged on the OpenAI API

The OpenAI Responses API now returns a 404 Not Found for "gpt-6-astra", where garbage/actually non-existent slugs return 400s - a 404 is also returned for 5.6 Cyber, which we know exists.

A similar quirk exists in the Bedrock API used to find Fable 5.1 a day prior to launch.

[Prabh @arora2992]

@alexgetmancom @thsottiaux 👀

[Prabh @arora2992]

@TokenGremlin "Token efficient" just means that it's your fault if your usage limits are going down even faster.

[Prabh @arora2992]

@TokenGremlin Can you add ->
-> Fucks the usage limits even on 20x even more than the "more token efficient" claimed 5.6 Sol

[Prabh @arora2992]

@thsottiaux @almmaasoglu I'm seeing things about my usage limits on 20x being fucked even more in the ass you wouldn't believe.

[Prabh @arora2992]

@almmaasoglu I'm hearing things about my usage limits on 20x being fucked even more in the ass you wouldn't believe

[Prabh @arora2992]

@kimmonismus Hmm.
So it doesn't beat Fable 5.1 and now they're delaying it?
Just tell us @kimmonismus I know you know! :P

[Prabh @arora2992]

@TokenGremlin The whole "token-efficient" marketing term translates in my mind to - "it's gonna eat your already reduced usage limits even more so just use it on low"

Same exact thing happened with 5.6 Sol

[Matthew Berman @MatthewBerman]

Already sent 25 codes ($5,000 worth!!)

Should I keep going?

How many will @SpaceXAI let me send? 😈

Let the tokens burn!

[Prabh @arora2992]

@MatthewBerman @SpaceXAI Me! I needs it!

[ChatGPT @ChatGPT]

And we're back! Get to Work.

[Prabh @arora2992]

@Redneck_Aladdin @Partici_Nation @TruthFairy131 Yup
Human trafficking is a big business
And the entire executive branch is in on it