trohrbaugh/Qwen3.8-27B-heretic-ara
#2798
by aoleg - opened
Please consider https://huggingface.co/trohrbaugh/Qwen3.8-27B-heretic-ara
Why: currently the best all-around abliteration of the new Qwen model:
Metric This model Original model (Qwen/Qwen3.8-27B)
KL divergence 0.0535 0 (by definition)
Refusals 0/100 99/100
queued with high priority before you asked hehe, but sadly nico1 is the only node that can process it and is sleeping. I wanted this model to try too, so it will be done first thing in the morning, but regardless, here's my usual:
It's queued! =)
You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#Qwen3.8-27B-heretic-ara-GGUF for quants to appear.