Honkware/Qwen3.8-27B-EfficientThink-Uncensored-K3-Opus5-Grok4.6-GPT5.6Sol-SFT-SimPO-DFlash2-exl3-4.0bpw

docs: proper model card with base-model link + inference notes

fix: coerce quantization_config.bits to int + add bits_per_weight

Upload README.md with huggingface_hub

Upload folder using huggingface_hub

initial commit