PrunaAI/Pruna-Qwen-Image-2.1
Adapter vs Qwen/Qwen-Image-2.1. PrunaAI/Pruna-Qwen-Image-2.1 is a LoRA adapter, not full weights: rank 64, on 0, gate_layer, out, proj, to_k, to_q, to_v (224 adapted matrices). It modifies Qwen/Qwen-Image-2.1 at load time.