> Are there any creative-writing LORAs published for open-weight models? Oddly enough standalone LoRA adapters are very popular in image generation world and

> Are there any creative-writing LORAs published for open-weight models? Oddly enough standalone LoRA adapters are very popular in image generation world and utterly unpopular in LLM world - there it is customary just to distribute fully merged models on HF. Having said that I personally use a finetune of Qwen 3.6, Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic, which was was not designed for creative writing per se but is far better than any other 24B-32B model I've tried for writing, let alone awful vanilla Qwen 3.6.