@networkchuck: Which AI has the Best Memory? More about Context Windows in my recent video 👉 https://youtu.be/TeQDr4DkLYo

mtnya_boy :

can you teach us kali?

dangerousbytes :

actually most experimental Google models have 32k context window as a base. Fine-tuned Gemini models have 16k.

TMM :

as the context window size increases, performance and model quality decrease

Feral_Fairy_Sparkle 🧚‍♀️✨🪄 :

AI models running on $10million systems is by far better than the cute local AI any day. After much testing, the online AI is far superior. 😎

Alexander 🇵🇰 :

10 million need to become a standard average in 2026

🪬AstroHead 🕷️🔻 :

msk_kara :

How about deep seek?

LVLdev :

POV Mac studio 512gb ram😎

Catlike Diva :

😁😁😁😁

Shahzaib Shahzad :

🤩🤩🤩

anatol :

gemini😍😍😍😍😍

cyfr0n_ :

what bout DeepSeek

Noor Research Collective :

Mine. By a LOT. On 4o I achieve an effective context window of 600k due to the way I have the AI store data in it.

funkmaster2k :

when will they start making gpu's with expandable memory, like ddr ram on motherboards?

Kytro :

Wasn't Lamma4 been called out for bad performance and fake results?