--- license: apache-2.0 pipeline_tag: image-to-video library_name: diffusers language: - es --- # Wan2.2-S2V-14B: Audio-Driven Cinematic Video Generation This repository features the **Wan2.2-S2V-14B** model, designed for audio-driven cinematic video generation. It was introduced in the paper: [**Wan-S2V: Audio-Driven Cinematic Video Generation**](https://huggingface.co/papers/2508.18621) ๐ Wan Homepage | ๐ฅ๏ธ GitHub | ๐ค Hugging Face Organization | ๐ค ModelScope Organization | ๐ Wan-S2V Paper | ๐ Wan2.2 Base Paper | ๐ Project Page | ๐ Blog | ๐ฌ Discord ๐ ไฝฟ็จๆๅ(ไธญๆ) | ๐ User Guide(English) | ๐ฌ WeChat(ๅพฎไฟก) ## Abstract (Wan-S2V Paper) Current state-of-the-art (SOTA) methods for audio-driven character animation demonstrate promising performance for scenarios primarily involving speech and singing. However, they often fall short in more complex film and television productions, which demand sophisticated elements such as nuanced character interactions, realistic body movements, and dynamic camera work. To address this long-standing challenge of achieving film-level character animation, we propose an audio-driven model, which we refere to as Wan-S2V, built upon Wan. Our model achieves significantly enhanced expressiveness and fidelity in cinematic contexts compared to existing approaches. We conducted extensive experiments, benchmarking our method against cutting-edge models such as Hunyuan-Avatar and Omnihuman. The experimental results consistently demonstrate that our approach significantly outperforms these existing solutions. Additionally, we explore the versatility of our method through its applications in long-form video generation and precise video lip-sync editing. ----- [**Wan: Open and Advanced Large-Scale Video Generative Models**](https://arxiv.org/abs/2503.20314) We are excited to introduce **Wan2.2**, a major upgrade to our foundational video models. With **Wan2.2**, we have focused on incorporating the following innovations: - ๐ **Effective MoE Architecture**: Wan2.2 introduces a Mixture-of-Experts (MoE) architecture into video diffusion models. By separating the denoising process cross timesteps with specialized powerful expert models, this enlarges the overall model capacity while maintaining the same computational cost. - ๐ **Cinematic-level Aesthetics**: Wan2.2 incorporates meticulously curated aesthetic data, complete with detailed labels for lighting, composition, contrast, color tone, and more. This allows for more precise and controllable cinematic style generation, facilitating the creation of videos with customizable aesthetic preferences. - ๐ **Complex Motion Generation**: Compared to Wan2.1, Wan2.2 is trained on a significantly larger data, with +65.6% more images and +83.2% more videos. This expansion notably enhances the model's generalization across multiple dimensions such as motions, semantics, and aesthetics, achieving TOP performance among all open-sourced and closed-sourced models. - ๐ **Efficient High-Definition Hybrid TI2V**: Wan2.2 open-sources a 5B model built with our advanced Wan2.2-VAE that achieves a compression ratio of **16ร16ร4**. This model supports both text-to-video and image-to-video generation at 720P resolution with 24fps and can also run on consumer-grade graphics cards like 4090. It is one of the fastest **720P@24fps** models currently available, capable of serving both the industrial and academic sectors simultaneously. ## Video Demos Your browser does not support the video tag. ## ๐ฅ Latest News!! * Aug 26, 2025: ๐ต We introduce **[Wan2.2-S2V-14B](https://humanaigc.github.io/wan-s2v-webpage)**, an audio-driven cinematic video generation model, including [inference code](#run-speech-to-video-generation), [model weights](#model-download), and [technical report](https://humanaigc.github.io/wan-s2v-webpage/content/wan-s2v.pdf)! Now you can try it on [wan.video](https://wan.video/), [ModelScope Gradio](https://www.modelscope.cn/studios/Wan-AI/Wan2.2-S2V) or [HuggingFace Gradio](https://huggingface.co/spaces/Wan-AI/Wan2.2-S2V)! * Jul 28, 2025: ๐ We have open a [HF space](https://huggingface.co/spaces/Wan-AI/Wan-2.2-5B) using the TI2V-5B model. Enjoy! * Jul 28, 2025: ๐ Wan2.2 has been integrated into ComfyUI ([CN](https://docs.comfy.org/zh-CN/tutorials/video/wan/wan2_2) | [EN](https://docs.comfy.org/tutorials/video/wan/wan2_2)). Enjoy! * Jul 28, 2025: ๐ Wan2.2's T2V, I2V and TI2V have been integrated into Diffusers ([T2V-A14B](https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B-Diffusers) | [I2V-A14B](https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B-Diffusers) | [TI2V-5B](https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B-Diffusers)). Feel free to give it a try! * Jul 28, 2025: ๐ We've released the inference code and model weights of **Wan2.2**. ## Community Works If your research or project builds upon [**Wan2.1**](https://github.com/Wan-Video/Wan2.1) or [**Wan2.2**](https://github.com/Wan-Video/Wan2.2), and you would like more people to see it, please inform us. - [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio) provides comprehensive support for Wan 2.2, including low-GPU-memory layer-by-layer offload, FP8 quantization, sequence parallelism, LoRA training, full training. - [Kijai's ComfyUI WanVideoWrapper](https://github.com/kijai/ComfyUI-WanVideoWrapper) is an alternative implementation of Wan models for ComfyUI. Thanks to its Wan-only focus, it's on the frontline of getting cutting edge optimizations and hot research features, which are often hard to integrate into ComfyUI quickly due to its more rigid structure. ## ๐ Todo List - Wan2.2-S2V Speech-to-Video - [x] Inference code of Wan2.2-S2V - [x] Checkpoints of Wan2.2-S2V-14B - [x] ComfyUI integration - [x] Diffusers integration ## Run Wan2.2 #### Installation Clone the repo: ```sh git clone https://github.com/Wan-Video/Wan2.2.git cd Wan2.2 ``` Install dependencies: ```sh # Ensure torch >= 2.4.0 # If the installation of `flash_attn` fails, try installing the other packages first and install `flash_attn` last pip install -r requirements.txt ``` #### Model Download | Models | Download Links | Description | |--------------------|---------------------------------------------------------------------------------------------------------------------------------------------|-------------| | T2V-A14B | ๐ค [Huggingface](https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B) ๐ค [ModelScope](https://modelscope.cn/models/Wan-AI/Wan2.2-T2V-A14B) | Text-to-Video MoE model, supports 480P & 720P | | I2V-A14B | ๐ค [Huggingface](https://huggingface.co/Wan-AI/Wan2.2-I2V-A14B) ๐ค [ModelScope](https://modelscope.cn/models/Wan-AI/Wan2.2-I2V-A14B) | Image-to-Video MoE model, supports 480P & 720P | | TI2V-5B | ๐ค [Huggingface](https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B) ๐ค [ModelScope](https://modelscope.cn/models/Wan-AI/Wan2.2-TI2V-5B) | High-compression VAE, T2V+I2V, supports 720P | | S2V-14B | ๐ค [Huggingface](https://huggingface.co/Wan-AI/Wan2.2-S2V-14B) ๐ค [ModelScope](https://modelscope.cn/models/Wan-AI/Wan2.2-S2V-14B) | Speech-to-Video model, supports 480P & 720P | Download models using huggingface-cli: ``` sh pip install "huggingface_hub[cli]" huggingface-cli download Wan-AI/Wan2.2-S2V-14B --local-dir ./Wan2.2-S2V-14B ``` Download models using modelscope-cli: ``` sh pip install modelscope modelscope download Wan-AI/Wan2.2-S2V-14B --local_dir ./Wan2.2-S2V-14B ``` #### Run Speech-to-Video Generation This repository supports the `Wan2.2-S2V-14B` Speech-to-Video model and can simultaneously support video generation at 480P and 720P resolutions. - Single-GPU Speech-to-Video inference ```sh python generate.py --task s2v-14B --size 1024*704 --ckpt_dir ./Wan2.2-S2V-14B/ --offload_model True --convert_model_dtype --prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard." --image "examples/i2v_input.JPG" --audio "examples/talk.wav" # Without setting --num_clip, the generated video length will automatically adjust based on the input audio length ``` > ๐ก This command can run on a GPU with at least 80GB VRAM. - Multi-GPU inference using FSDP + DeepSpeed Ulysses ```sh torchrun --nproc_per_node=8 generate.py --task s2v-14B --size 1024*704 --ckpt_dir ./Wan2.2-S2V-14B/ --dit_fsdp --t5_fsdp --ulysses_size 8 --prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard." --image "examples/i2v_input.JPG" --audio "examples/talk.wav" ``` - Pose + Audio driven generation ```sh torchrun --nproc_per_node=8 generate.py --task s2v-14B --size 1024*704 --ckpt_dir ./Wan2.2-S2V-14B/ --dit_fsdp --t5_fsdp --ulysses_size 8 --prompt "a person is singing" --image "examples/pose.png" --audio "examples/sing.MP3" --pose_video "./examples/pose.mp4" ``` > ๐กFor the Speech-to-Video task, the `size` parameter represents the area of the generated video, with the aspect ratio following that of the original input image. > ๐กThe model can generate videos from audio input combined with reference image and optional text prompt. > ๐กThe `--pose_video` parameter enables pose-driven generation, allowing the model to follow specific pose sequences while generating videos synchronized with audio input. > ๐กThe `--num_clip` parameter controls the number of video clips generated, useful for quick preview with shorter generation time. ## Computational Efficiency on Different GPUs We test the computational efficiency of different **Wan2.2** models on different GPUs in the following table. The results are presented in the format: **Total time (s) / peak GPU memory (GB)**. > The parameter settings for the tests presented in this table are as follows: > (1) Multi-GPU: 14B: `--ulysses_size 4/8 --dit_fsdp --t5_fsdp`, 5B: `--ulysses_size 4/8 --offload_model True --convert_model_dtype --t5_cpu`; Single-GPU: 14B: `--offload_model True --convert_model_dtype`, 5B: `--offload_model True --convert_model_dtype --t5_cpu` (--convert_model_dtype converts model parameter types to config.param_dtype); > (2) The distributed testing utilizes the built-in FSDP and Ulysses implementations, with FlashAttention3 deployed on Hopper architecture GPUs; > (3) Tests were run without the `--use_prompt_extend` flag; > (4) Reported results are the average of multiple samples taken after the warm-up phase. ------- ## Introduction of Wan2.2 **Wan2.2** builds on the foundation of Wan2.1 with notable improvements in generation quality and model capability. This upgrade is driven by a series of key technical innovations, mainly including the Mixture-of-Experts (MoE) architecture, upgraded training data, and high-compression video generation. ##### (1) Mixture-of-Experts (MoE) Architecture Wan2.2 introduces Mixture-of-Experts (MoE) architecture into the video generation diffusion model. MoE has been widely validated in large language models as an efficient approach to increase total model parameters while keeping inference cost nearly unchanged. In Wan2.2, the A14B model series adopts a two-expert design tailored to the denoising process of diffusion models: a high-noise expert for the early stages, focusing on overall layout; and a low-noise expert for the later stages, refining video details. Each expert model has about 14B parameters, resulting in a total of 27B parameters but only 14B active parameters per step, keeping inference computation and GPU memory nearly unchanged. The transition point between the two experts is determined by the signal-to-noise ratio (SNR), a metric that decreases monotonically as the denoising step $t$ increases. At the beginning of the denoising process, $t$ is large and the noise level is high, so the SNR is at its minimum, denoted as ${SNR}_{min}$. In this stage, the high-noise expert is activated. We define a threshold step ${t}_{moe}$ corresponding to half of the ${SNR}_{min}$, and switch to the low-noise expert when $t To validate the effectiveness of the MoE architecture, four settings are compared based on their validation...
