Hacker News new | ask | show | jobs
by cubefox 530 days ago
There were multiple reports confirming that OpenAI's Orion (planned to be GPT-5) yielded unexpectedly weak results.
3 comments

And not just OpenAI is facing this problem. Anthropic and Google as well.
So Deepseek V3 did nothing to show you how wrong this take is?
And costs $500 million per training run.