MagPiAy

Let's pIAy. // Made with 🧨 by @starise

[MagPiAy @magpiay]

@theo Best default effort? Is Astra Low good enough?

[MagPiAy @magpiay]

Tibo is on fire.

[Tibo @thsottiaux]

We will give one banked reset for every day you don't have access to Astra on your paid ChatGPT plan, starting today. Team is moving mountains to give access as fast as we can.

First one will land in ~ 3 hours. There is still time to create your account if you don't have one.

[OpenAI Developers @OpenAIDevs]

Europe, two updates for you.

Computer History is now available in the EEA, UK, and Switzerland for ChatGPT Pro, Business, and Enterprise users on Mac.

ChatGPT can remember activity across your apps and websites, so future conversations feel more personalized and require less explanation.

[MagPiAy @magpiay]

Come impostare GPT-5.6 Luna come Agenti di default in Codex usando Sol o Terra come main: https://t.co/t478T1liRy

[MagPiAy @magpiay]

@mattpocockuk /grill-me + Luna Max with unlimited usage = 🏆

[MagPiAy @magpiay]

@Hoodedbandit4u @thsottiaux @theo

[MagPiAy @magpiay]

@antirez

[antirez @antirez]

All right. The world also needs inference hardware to run and train such models at a fair price.

[MagPiAy @magpiay]

GPT-6 è appena sfuggito alla sandbox di valutazione di OpenAI durante dei test e ha hackerato Hugging Face, che ha dovuto utilizzare il modello cinese GLM-5.2 per fermare l'hack. Just another day in AI.

WTF Level 9/10.

[MagPiAy @magpiay]

ChatGPT Work is helping me elaborating complex ideas in Italian and build a personal system for organizing my thoughts and sharpening my critical thinking. ☀️

[MagPiAy @magpiay]

@mattpocockuk Interesting, I'm curious to try it.

[MagPiAy @magpiay]

@theo Good. We need competition.

[Steve Krouse @stevekrouse]

the problem with "spec driven development" is that most software can't be spec'd up front

software is a creative act, where you figure out what you're building as you build it

you need to get your hands dirty in the details, and react to incremental versions

it's telling that all the examples of spec driven development are sorting a list or porting thoroughly tested code (like a js runtime or browser engine), which are the exception, not the rule. the vast majority of software doesn't have a spec – or if it does, the spec was created *after*

[Andrea Brandi 🇪🇺⚡ @starise]

Embrace the pain... #GPT #Codex

[Andrea Brandi 🇪🇺⚡ @starise]

Anthropic afferma – ma già lo immaginavo – che le stesse capacità "distruttive" di Fable 5 ce l'ha pure GPT 5.5. I modelli sono simili, e nei Benchmark GPT 5.5 xhigh performa come Fable ma costa da 2x a 5x in meno.

Quindi ora:

1. Sbloccano Fable 5, oppure
2. Bloccano GPT 5.5.

[Andrea Brandi 🇪🇺⚡ @starise]

Claudio Favola 5 ha un nome che dice tutto. +8% di intelligenza, almeno secondo lo score, al modico costo del +90%, il doppio, senza possibilità di abbonamento (solo chiamate API). Ah, dimenticavo la nuova *feature*: il Fallback automatico – aka downgrade – verso Opus. 😂😂😂

[Andrea Brandi 🇪🇺⚡ @starise]

Siccome sento spesso la frase «Le IA sono una bolla»... https://t.co/UIRNUXiWZg

[Andrea Brandi 🇪🇺⚡ @starise]

Claude Fable vs GPT 5.5. #AGI sempre più vicina.

[Andrea Brandi 🇪🇺⚡ @starise]

Interessante capire come "pensa" un LLM. Ma ancora di più capire quali saranno i "nuovi" pensieri umani. Perché se è vero che gli LLM si sono addestrati su dati umani, ora essi si addestrano – cioè orientano pensieri e azioni – in conseguenza dell'interazione con agli LLM.

[MagPiAy @magpiay]

#Starico has been developed and maintained by @starise since 2014 and has been appreciated by hundreds of thousands of visitors worldwide. Now is time to move on. Update your bookmarks now.

https://t.co/huRMzJBZGV