UNIT.City — місце, де люди працюють... КРАЩЕ! Обирай свій простір просто зараз 👉
Олександр КузьменкоAI Eng
4 March 2025, 10:48
2025-03-04
A shortage of graphics chips forced OpenAI to suspend the deployment of the new GPT-4.5 model
OpenAI CEO Sam Altman said the company was forced to pause the release of its latest artificial intelligence model, GPT-4.5, because it «ran out of GPUs.» As a result, only subscribers to the most expensive $200 Pro subscription have access to GPT-4.5, but the situation may improve as early as this week.
OpenAI CEO Sam Altman said the company was forced to pause the release of its latest artificial intelligence model, GPT-4.5, because it «ran out of GPUs.» As a result, only subscribers to the most expensive $200 Pro subscription have access to GPT-4.5, but the situation may improve as early as this week.
OpenAI recently released its latest model, GPT-4.5. However, it is currently only available to Pro subscribers who pay a $200 per month subscription.
The company’s CEO Sam Altman said that he was forced to delay the release of the model because «…we are growing very quickly, and we are running out of GPUs.» According to him, this week it will add «tens of thousands of GPUs» and the model will become available to Plus subscribers for $20.
Altman added in his post that hundreds of thousands more GPUs are coming soon. That shortage is forcing OpenAI to develop its own AI silicon in partnership with Broadcom. But since it will be years before the company can release its own chips, it must rely on Nvidia and other suppliers for its needs for now.
Altman noted that GPT-4.5 «is a giant, expensive model.» It costs $75 per million input tokens and $150 per million output tokens. By comparison, GPT-4o costs just $2.50 per million input tokens and $10 per million output tokens. Despite the hefty price tag, Altman says it’s not a reasoning model and won’t set benchmark records. But he added that «it’s a different kind of intelligence, and there’s a magic to it» that he hasn’t experienced before.
Recall that GPT-4.5 was created with scalable learning methods in mind, which allowed the model to become even more reliable and accurate. GPT-4.5 gives 24,3% more accurate answers than GPT-4o. The new model is better at following user queries, which makes it effective for writing texts, coding, analytics, and other tasks.
Microsoft previously announced that all Copilot users will get unlimited access to Think Deeper and Voice at no additional cost. These two features can be considered analogs of ChatGPT Advanced Voice and o1, for which ChatGPT Plus users pay $20 per month. So this is an opportunity to try them for free in the Copilot version.
Як нейромережі бачать вільну та незалежну Україну? Тест dev.ua
Нейронні мережі для генерації зображень бачать світ по-своєму, їхню логіку зрозуміти часом зовсім неможливо. Але таки хочеться. На честь Дня Незалежності України редакція dev.ua вирішила провести невеликий експеримент.
Ми задали чотирьом різним нейронним мережам п’ять однакових запитів: «прапор України», «День Незалежності України», «український Крим», «перемога України» та «українці». Отриманими результатами ми ділимося з вами нижче.
У TikTok тепер можна генерувати фон за допомогою нейромережі. Ми протестували її та ділимося результатами
У TikTok з’явилася нова функція «Розумний фон». З її допомогою як фон для тіктоків можна підставляти згенеровані нейромережею зображення. Редакція dev.ua протестувала цю технологію і ділиться своїми враженнями.