open-source-lm

LuffyTheFox released a GGUF quantized version of the Qwen3.6-35B-A3B model, denoised with the Genesis algorithm to reduce tensor noise and improve instruction following, based on an uncensored aggressive variant merged with Hermes finetune data.