rockerBOO/qwen3-4b-roleplay-lora-F16-GGUF
Adapter vs rockerBOO/qwen3-4b-roleplay-lora. rockerBOO/qwen3-4b-roleplay-lora is a LoRA adapter, not full weights: rank 8, on k_proj, o_proj, q_proj, v_proj in 36 layers (144 adapted matrices). It modifies rockerBOO/qwen3-4b-roleplay-lora-F16-GGUF at load time.