Wan2.1_i2v_720p_14b_fp16.safetensors
14 Billion parameters, which allows it to capture complex prompts and subtle facial/environmental movements more accurately than the lightweight 1.3B model. Why This Model Matters Wan-Video/Wan2.1 - GitHub
14 Billion parameters, which allows it to capture complex prompts and subtle facial/environmental movements more accurately than the lightweight 1.3B model. Why This Model Matters Wan-Video/Wan2.1 - GitHub