Qwen3.8 Goes UncensoredA modified Qwen3.8 27B-Uncensored-FP8 model has appeared in open source with much fewer built-in restrictions. It is aimed at developers who want to study LLM behavior without additional filters.
The model is available on Hugging Face and is intended for experiments, research, and discussion of complex or controversial topics.
Almost no built-in limits on topics and requests More open discussion of complex and controversial subjects Expanded options for experiments and research 27B model in Uncensored-FP8 format
Link 
Use it responsibly and consider where and how the model should be applied.
@program
