←back to thread

GPT-5.2

(openai.com)
1019 points atgctg | 1 comments | 11 Dec 25 18:04 UTC | HN request time: 0s | source
Show context
MagicMoonlight ◴[11 Dec 25 18:59 UTC] No.46235544[source]▶
They’re definitely just training the models on the benchmarks at this point
replies(1): >>46235649 #
1. roxolotl ◴[11 Dec 25 19:05 UTC] No.46235649[source]▶
Yea either this is an incredible jump or we’ve finally gotten confirmation benchmarks are bs.