alt Hacker NewsAs we saw with GPT-5 the RL technique of training doesn't scale forever
Replies
oezi • last Sunday at 9:19 PM
I meant scaling the base training before RL.
As we saw with GPT-5 the RL technique of training doesn't scale forever
I meant scaling the base training before RL.
Unless GPT-5 is 30% cheaper to run than o3. Then it's scaling brilliantly given the small gap between release dates. People are really drawing too many conclusions from too little information.