Viewing profile — kisjovan

You should check out our TerminalBench2.1 score for that, the tasks there can run up to 4h! We got almost no loss (we got Base 66.74% vs Swift 65.84%) with -38.7% thinking token re…

Thank you so much!! It would be great if you could share some numbers with the community :)

I will TLDR you on our thought process, research, training and benchmarks. 1. When running our quantized Qwen 3.8 27B instances we were very annoyed by random reasoning loops (in t…

Hi everybody, we post-trained Qwen 3.8 27B to be more efficient by figuring out which tokens were linked to overthinking and penalizing them without "attacking" the reasoning lengt…