## Overview Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v is an advanced image-to-video generation model built upon the Wan2.1-I2V-14B-480P foundation. This approach allows the model to generate videos with significantly fewer inference steps (4 steps) and without classifier-free guidance, substantially reducing video generation time while maintaining high quality outputs. In this version, we added the following features: 1. Trained with higher quality datasets for extended iterations. 2. New fp8 and int8 quantized distillation models have been added, which enable fast inference using lightx2v on RTX 4060. ## Training Our training code is modified based on the [Self-Forcing](https://github.com/guandeh17/Self-Forcing) repository. We extended support for the Wan2.1-14B-I2V-480P model and performed a 4-step bidirectional distillation process. The modified code is available at [Self-Forcing-Plus](https://github.com/GoatWu/Self-Forcing-Plus). ## Inference Our inference framework utilizes [lightx2v](https://github.com/ModelTC/lightx2v), a highly efficient inference engine that supports multiple models. This framework significantly accelerates the video generation process while maintaining high quality output. Before running either script, set `lightx2v_path`, `model_path`, `CUDA_VISIBLE_DEVICES`, and the input image/output paths. Keep `model_path` pointed at the complete Wan2.1-I2V-14B-480P base model directory. Run the commands from the LightX2V repository root. **Full distilled model** In `configs/distill/wan21/wan_i2v_distill_model_4step_cfg.json`, set `dit_original_ckpt` to the local `distill_models/distill_model.safetensors` file downloaded from this repository. ```bash bash scripts/wan/inference/distill/run_wan_i2v_distill_model.sh ``` **LoRA** In `configs/distill/wan21/wan_i2v_distill_lora_4step_cfg.json`, set `lora_configs[0].path` to the local `loras/Wan21_I2V_14B_lightx2v_cfg_step_distill_lora_rank64.safetensors` file downloaded from this repository. ```bash bash scripts/wan/inference/distill/run_wan_i2v_distill_lora.sh ``` We recommend using the **LCM scheduler** with the following settings: - `shift=5.0` - `guidance_scale=1.0 (i.e., without CFG)` ## License Agreement The models in this repository are licensed under the Apache 2.0 License. We claim no rights over the your generate contents, granting you the freedom to use them while ensuring that your usage complies with the provisions of this license. You are fully accountable for your use of the models, which must not involve sharing any content that violates applicable laws, causes harm to individuals or groups, disseminates personal information intended for harm, spreads misinformation, or targets vulnerable populations. For a complete list of restrictions and details regarding your rights, please refer to the full text of the [license](LICENSE.txt). ## Acknowledgements We would like to thank the contributors to the [Wan2.1](https://huggingface.co/Wan-AI/Wan2.1-T2V-14B), [Self-Forcing](https://huggingface.co/gdhe17/Self-Forcing/tree/main) repositories, for their open research.