Tips

Our model is still in the process of iterative upgrades, and we will open it up for user testing within the next 2-3 months.
At present, we welcome everyone to send us the samples they want to test. As long as suitable instructions are provided, we believe that everyone can create videos with cooler effects：

  • The instructions should be in the format: <Text description, First frame guidance (image), Last frame guidance (image)> or <Text description, First frame guidance (image)>;
    • Please note that it is hard for the model to understand particularly complex text. Complex video effects may need to rely on information provided by the <Last frame>. The text description mainly plays a supporting role;
    • the text description only supports English;
  • We will prioritize processing cooler test samples;
  • Please do not send pornographic, political, bloody, etc. test samples;
  • Please indicate in the email whether we are allowed to publicly display it. We will prioritize cases that allow public display;
  • The model effect is still being iteratively improved. There is limited manpower for testing, please be patient.
  • Contact: zjn@bytedance.com

Tips

我们的模型仍在迭代升级中，我们将会在2-3月内开放用户测试。
目前，我们欢迎大家把想要测试的样例发给我们，只要提供合适的instructions，我们相信大家一定能做出更酷炫的效果：

  • instructions格式为：<描述(纯文本), 首帧指导(图片), 尾帧指导(图片)>，或者<描述(纯文本), 首帧指导(图片)>；
    • 注意不要期待模型很好地理解特别复杂的文本，复杂的视频效果需要依靠<尾帧>提供信息，文本描述主要起到配合作用；
    • 文本描述仅支持英文；
  • 我们会优先处理更酷炫的测试样例；
  • 请不要发送 色情、政治、血腥等测试样例;
  • 请在邮件中注明是否允许我们公开展示，我们会优先处理允许公开的样例；
  • 模型效果仍在迭代提升中，用于测试的人力有限，请耐心等待。
  • 联系方式：zjn@bytedance.com