A $500 RL fine-tune of a 9B open model beat frontier models on catalog review

Re:A $500 RL fine-tune of a 9B open model beat frontier mode... by fnands on Tuesday July 28, 2026 @03:50AM (#49080717)

Data drift. The data you are doing real world inference on can start changing, meaning your model performance starts degrading, so the model needs to be retrained on new data (that you have to collect and prepare).

[ Parent | on HN ]

Re:A $500 RL fine-tune of a 9B open model beat frontier mode... by Turskarama on Tuesday July 28, 2026 @04:31AM (#49081030)

Sure, but that apparently only costs $500. If you do that once a month then that's still basically free. Hell if you do it once a week it's still about as much as a single cheap employee.

[ Parent | on HN ]

Re:A $500 RL fine-tune of a 9B open model beat frontier mode... by zem on Tuesday July 28, 2026 @05:43AM (#49081542)

you missed the parent comment's point that the $500 training run was merely the last step, generating the training data is likely to be significantly more expensive and time-consuming. might still pay for itself, but it's not a trivial "spend $500 every month" decision.

[ Parent | on HN ]

Re:A $500 RL fine-tune of a 9B open model beat frontier mode... by Foobar8568 on Tuesday July 28, 2026 @09:14AM (#49083350)

My company wanted me to help them to fine tune a model. I asked them if I were working full time on it, never heard back. $500 is largely the cheapest part of it, but I know from experience, most people are clueless about data.

I worked for a billion+ revenue business where they didn't realize one of their key data hasn't been updated for 2 years, until I join them, and these types of screw up is cross business/functions/entities.

[ Parent | on HN ]