Re:GPT-6 Astra, Looped Transformers, and Hidden Reasoning
by vatsachak on Wednesday September 09, 2026 @06:59PM
(#49635788)
What was your prompt? I have gotten easy fixes with I tell if to do something.
Re:GPT-6 Astra, Looped Transformers, and Hidden Reasoning
by jiggawatts on Thursday September 10, 2026 @05:15AM
(#49640755)
Try Astra on low or at most medium thinking level. Its "low" is better than Sol "high" or even "xhigh", and then it also doesn't overthink as much.
It seems to overengineer really bad and it is also very slow due to it "thinking" too much I feel like.
One example is that I asked it to implement a new functionality inside an existing App of mine and if I had written it myself it would have been like a ~50 line diff. Astra took like 10 minutes to write ~400 lines, most of them useless and also in pretty bad style, barely readable code.
Maybe I am bad with prompting but I didn't have these issues before, not even with 5.6 Sol on max reasoning.