@1littlecoder ARC-AGI-2 score is also bonkers

@Prince_Canuma crazy numbers

@1littlecoder They used too much thinking compute for those benchmarks, not the kind you will do on average tasks.

@1littlecoder 😏

@1littlecoder Ive kinda learned to wait for third-party benchmarks. Opus 4.5 was supposed to be so much, according to benchmarks, but it was just an expert coder. (Which isnt knocking it.) I have my sites for trusted benchmarks, and I look to those.

@1littlecoder Sounds totally legit to me. What scam?

More than 60% of workers are forced to wait 400 days or longer just to get a first contract after they vote to unionize. This is corporate corruption. The Faster Labor Contracts Act would make fast first contracts a reality for millions of American workers — and the Teamsters

Realtime speech to speech translation powered by Gemini, available in Google Translate now, coming to developers early next year : ) https://t.co/3O3ovULHXe

GPT-5.2 exceeded a trillion tokens in the API on its first day of availability and is growing fast!

I don’t get it: why would anyone pay thousands of $ to consulting companies or people when SOTA AI models like GPT-5.2 Pro, Gemini Deep Research & NotebookLM are better than the top 1% of experts in any field? For few hundred $ a month, AI is the best consulting in the world!

Looks like @OpenAI has added an even MORE powerful version of Pro mode... you can now ask GPT-5.2 Pro to think even longer than before. Starting to test this... I have high expectations here.