## Basic model info Model name: wan-video/wan-2.2-i2v-a14b Model description: Image-to-video at 720p and 480p with Wan 2.2 A14B ## Model inputs - prompt (required): Prompt for video generation (string) - image (optional): Input image to generate video from. (string) - go_fast (optional): Go fast (boolean) - num_frames (optional): Number of video frames. 81 frames give the best results (integer) - resolution (optional): Resolution of video. 832x480px corresponds to 16:9 aspect ratio, and 480x832px is 9:16 (string) - frames_per_second (optional): Frames per second. Note that the pricing of this model is based on the video duration at 16 fps (integer) - sample_steps (optional): Number of generation steps. Fewer steps means faster generation, at the expensive of output quality. 30 steps is sufficient for most prompts (integer) - sample_shift (optional): Sample shift factor (number) - seed (optional): Random seed. Leave blank for random (integer) ## Model output schema { "type": "string", "title": "Output", "format": "uri" } If the input or output schema includes a format of URI, it is referring to a file. ## Example inputs and outputs Use these example outputs to better understand the types of inputs the model accepts, and the types of outputs the model returns: ### Example (https://replicate.com/p/m9655vk2bdrma0craxy9q422eg) #### Input ```json { "image": "https://replicate.delivery/pbxt/NRZUpUbchqkonq8vJyY9MhE4CdkmsxNyoDyGsx8IWjmm0W3S/i2v_input.JPG", "prompt": "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside", "num_frames": 81, "resolution": "480p", "sample_shift": 5, "sample_steps": 30, "frames_per_second": 16 } ``` #### Output ```json "https://replicate.delivery/xezq/7n4vozePFawDDCPgfefvnhhVrZrLoZHRXiK14vcvPwdu7xWUB/output.mp4" ``` ### Example (https://replicate.com/p/43a2hdchjxrma0crayabpygmdc) #### Input ```json { "image": "https://replicate.delivery/pbxt/NRZtrR3lUtABKDbQ1E70AdzjuBXhLd1WJ8EovYI9wPQRXcfl/gondola.jpg", "prompt": "Golden hour, soft lighting, warm colors, saturated colors, wide shot, left-heavy composition. A weathered gondolier stands in a flat-bottomed boat, propelling it forward with a long wooden pole through the flooded ruins of Venice. The decaying buildings on either side are cloaked in creeping vines and marked by rusted metalwork, their once-proud facades now crumbling into the water. The camera moves slowly forward and tilts left, revealing behind him the majestic remnants of the city bathed in the amber glow of the setting sun. Silhouettes of collapsed archways and broken domes rise against the golden skyline, while the still water reflects the warm hues of the sky and surrounding structures.", "num_frames": 81, "resolution": "480p", "sample_shift": 5, "sample_steps": 30, "frames_per_second": 16 } ``` #### Output ```json "https://replicate.delivery/xezq/IFCJhDosHSqqEhcKkcEBHsyqtZOkvnfNYfTTub8C9Ryn3sFVA/output.mp4" ``` ### Example (https://replicate.com/p/dg0hhe2xzxrmc0craycb99pa98) #### Input ```json { "image": "https://replicate.delivery/pbxt/NRZxmfAJxTeccKutQee0zGEgkDcsfaMUdczPA5p6kx3iLErA/eye.jpg", "prompt": "The reflection of someone is in her glassy eye as she blinks. Her eyes capture her awe.", "num_frames": 81, "resolution": "480p", "sample_shift": 5, "sample_steps": 30, "frames_per_second": 16 } ``` #### Output ```json "https://replicate.delivery/xezq/ioJSIeWopn0daqzF05dq2A6yoRIbodaZEtf4poZdb2gg7sFVA/output.mp4" ``` ### Example (https://replicate.com/p/25y1wfd5jxrme0craye9w344hc) #### Input ```json { "image": "https://replicate.delivery/pbxt/NRa2Ef9CJ9FI7VtmKWKskgHLfCsMPOK1dJlGYUZawjSZwrNz/4b9d68c0-09ce-4fda-8b76-21a14342d088.jpg", "prompt": "They are biking", "num_frames": 81, "resolution": "720p", "sample_shift": 5, "sample_steps": 30, "frames_per_second": 16 } ``` #### Output ```json "https://replicate.delivery/xezq/Ctposa2BPYppCJ30xX7T1SRpIGouaI8b8zZXvG9zujvKQbRF/output.mp4" ``` ### Example (https://replicate.com/p/fv5w2graqnrme0craygsh5z3yc) #### Input ```json { "image": "https://replicate.delivery/pbxt/NRa6l5eoS23yD52EIj7dCifsx7OLmDLRCCXy0NglSGdBvcx0/tmpamy1gm7s.png", "prompt": "She takes off her helmet and looks around", "num_frames": 81, "resolution": "720p", "sample_shift": 5, "sample_steps": 30, "frames_per_second": 16 } ``` #### Output ```json "https://replicate.delivery/xezq/chxPBMaAYI65IlGEcqPe11WHgIuHb2V7xfbb17oXDijNFtFVA/output.mp4" ``` ## Model readme > ## 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 $t<{t}_{moe}$. > > > > > > To validate the effectiveness of the MoE architecture, four settings are compared based on their validation loss curves. The baseline **Wan2.1** model does not employ the MoE architecture. Among the MoE-based variants, the **Wan2.1 & High-Noise Expert** reuses the Wan2.1 model as the low-noise expert while uses the Wan2.2's high-noise expert, while the **Wan2.1 & Low-Noise Expert** uses Wan2.1 as the high-noise expert and employ the Wan2.2's low-noise expert. The **Wan2.2 (MoE)** (our final version) achieves the lowest validation loss, indicating that its generated video distribution is closest to ground-truth and exhibits superior convergence. > > > ##### (2) Efficient High-Definition Hybrid TI2V > To enable more efficient deployment, Wan2.2 also explores a high-compression design. In addition to the 27B MoE models, a 5B dense model, i.e., TI2V-5B, is released. It is supported by a high-compression Wan2.2-VAE, which achieves a $T\times H\times W$ compression ratio of $4\times16\times16$, increasing the overall compression rate to 64 while maintaining high-quality video reconstruction. With an additional patchification layer, the total compression ratio of TI2V-5B reaches $4\times32\times32$. Without specific optimization, TI2V-5B can generate a 5-second 720P video in under 9 minutes on a single consumer-grade GPU, ranking among the fastest 720P@24fps video generation models. This model also natively supports both text-to-video and image-to-video tasks within a single unified framework, covering both academic research and practical applications. > > > > > > > > > ##### Comparisons to SOTAs > We compared Wan2.2 with leading closed-source commercial models on our new Wan-Bench 2.0, evaluating performance across multiple crucial dimensions. The results demonstrate that Wan2.2 achieves superior performance compared to these leading models. > > > > > > ## Citation > If you find our work helpful, please cite us. > > ``` > @article{wan2025, > title={Wan: Open and Advanced Large-Scale Video Generative Models}, > author={Team Wan and Ang Wang and Baole Ai and Bin Wen and Chaojie Mao and Chen-Wei Xie and Di Chen and Feiwu Yu and Haiming Zhao and Jianxiao Yang and Jianyuan Zeng and Jiayu Wang and Jingfeng Zhang and Jingren Zhou and Jinkai Wang and Jixuan Chen and Kai Zhu and Kang Zhao and Keyu Yan and Lianghua Huang and Mengyang Feng and Ningyi Zhang and Pandeng Li and Pingyu Wu and Ruihang Chu and Ruili Feng and Shiwei Zhang and Siyang Sun and Tao Fang and Tianxing Wang and Tianyi Gui and Tingyu Weng and Tong Shen and Wei Lin and Wei Wang and Wei Wang and Wenmeng Zhou and Wente Wang and Wenting Shen and Wenyuan Yu and Xianzhong Shi and Xiaoming Huang and Xin Xu and Yan Kou and Yangyu Lv and Yifei Li and Yijing Liu and Yiming Wang and Yingya Zhang and Yitong Huang and Yong Li and You Wu and Yu Liu and Yulin Pan and Yun Zheng and Yuntao Hong and Yupeng Shi and Yutong Feng and Zeyinzi Jiang and Zhen Han and Zhi-Fan Wu and Ziyu Liu}, > journal = {arXiv preprint arXiv:2503.20314}, > year={2025} > } > ``` > > ## License Agreement > The models in this repository are licensed under the Apache 2.0 License. We claim no rights over the your generated 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).