Topics
Latest
AI
Amazon
Image Credits:Eugene Gologursky / Getty Images for the New York Times
Apps
Biotech & Health
mood
Image Credits:Eugene Gologursky / Getty Images for the New York Times
Cloud Computing
commercialism
Crypto
Enterprise
EVs
Fintech
Fundraising
Gadgets
Gaming
Government & Policy
Hardware
Layoffs
Media & Entertainment
Meta
Microsoft
privateness
Robotics
Security
societal
infinite
startup
TikTok
Transportation
speculation
More from TechCrunch
issue
Startup Battlefield
StrictlyVC
Podcasts
Videos
Partner Content
TechCrunch Brand Studio
Crunchboard
touch Us
OpenAI CEO Sam Altman allege that the company was forced to careen the rollout of its newest model , GPT-4.5 , because OpenAI is “ out of GPUs . ”
Perhaps in part due to its enormous size , GPT-4.5 is wildly expensive . OpenAI is charge $ 75 per million tokens ( ~750,000 Bible ) feed into the role model and $ 150 per million tokens generate by the model . That ’s 30x the input price and 15x the output cost of OpenAI ’s workhorseGPT-4omodel .
GPT 4.5 pricing is disquiet . If this does n’t have enormous models smack , I will be disappointedpic.twitter.com/1kK5LPN9GH
— Casper Hansen ( @casper_hansen_)February 27 , 2025
“ We ’ve been acquire a caboodle and are out of GPUs , ” Altman wrote . “ We will add tens of thousands of GPUs next hebdomad and cast it out to the Plus tier then … This is n’t how we want to operate , but it ’s hard to perfectly predict growth surges that chair to GPU dearth . ”
Altman haspreviouslysaid that a want of computing capacity is stay the company ’s products . OpenAI hopes to combat this in the get age bydeveloping its own AI chipsand bybuilding a massive web of information centers .
