FroxStudio/Flex-Model-LoRA
Adapter vs Qwen/Qwen2.5-1.5B-Instruct. FroxStudio/Flex-Model-LoRA is a LoRA adapter, not full weights: rank 16, on down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj in 28 layers (196 adapted matrices). It modifies Qwen/Qwen2.5-1.5B-Instruct at load time.