News for hackers. Stuff that matters.

A comment on OpenAI: GPT 5.6 Sol price reduction (until at least Nov 21). Parent comment

Re:OpenAI: GPT 5.6 Sol price reduction (until at least Nov 21) by colingauvin on Monday August 24, 2026 @08:41PM (#49427745)
I'm supposed to believe that DeepSeek distilled a 300B-1T param model with 150,000 requests? Lol.
[ Parent | on HN ]
Re:OpenAI: GPT 5.6 Sol price reduction (until at least Nov 21) by brookst on Monday August 24, 2026 @11:47PM (#49428849)
Are you thinking distillation goes from zero to complete model?

I believe it can be used in the RL / fine tuning sense, in which case 150,000 requests, assuming every one was detected, could move the needle in quality.

I agree it couldn’t replace all of pre and post training , but I don’t think that’s the claim. You do typical training, then distill really difficult cases.

[ Parent | on HN ]