Ali Tongyi 10,000-phase video generation model "AI raw video" function is officially online, more understanding of the Chinese style of the big model is here!

Ali Tongyi 10,000-phase video generation model "AI raw video" function is officially online, more understanding of the Chinese style of the big model is here!

At the 2024 Ali Yunqi Conference, AliCloud CTO Zhou Jingren announced that its newly developed AI video generation grand model-Tongyi WanxiangAI Raw Video, is now officially live and available to try out right away on both the official website and the app.

AI video domestic battlefield, Ali also down.

Alibaba's Tongyi launched the Tongyi 10,000 phase AI born video model was officially launched, with a strong picture visual dynamic generation capabilities, support for a variety of artistic styles and film and television level texture of video content generation.

The model optimizes the expression of Chinese elements, supports multi-language input and variable resolution generation, has a wide range of application scenarios, provides free services and features audio generation to simplify the video production process.

Tongyi's AI video model has a powerful visual dynamic generation capability, which supports the generation of video content with various artistic styles and movie-level textures.

Optimized Chinese style elements performance, with unique advantages in generating national style content, supporting multi-language input and variable resolution generation to meet different user needs.

Provides a free service that supports audio generation of video content, simplifies the video production process, synchronizes audio and video, and improves creative efficiency.

Compared to Sora and Gen-3 Alpha, which have exploded abroad, Tongyi Wanxiang is an AI video model that understands Chinese language and Chinese style better.

It is capable of supporting up to 5 seconds of video generation at 30 frames per second with a resolution of 720 P. Even more amazing is that it can also generate sound effects to match the picture.

This is backed by Ali's fully self-developed visual big model, and uses the industry-leading core architecture - Diffusion+Transformer.

Highlights: mobile app unlimited times, completely free to use! PC, daily login to send 50 inspiration value, can be generated 10 times.

AttachedExperience Links:

https://tongyi.aliyun.com/wanxiang/wanxvideo

statement:The content is collected from various media platforms such as public websites. If the included content infringes on your rights, please contact us by email and we will deal with it as soon as possible.
Information

NVIDIA Predicts "GPT-3 Moment" in Robotics in the Next 2-3 Years as "Terminator" Prototype Comes to Life

2024-9-20 10:02:04

Information

OpenAI's new AI models o1-preview and o1-mini achieve top scores in chatbot rankings

2024-9-21 10:45:21

Search