Alibaba, Metinlerden Resim ve Video Oluşturan Üst Düzey Yapay Zekâ Araçını Ücretsiz Olarak Tanıttı!
Recently, Chinese e-commerce giant Alibaba has made significant strides in the AI sector with the announcement of its new AI model “Wan 2.1.” This tool can generate photos and videos from text, attracting attention primarily for its open-source and free availability.
According to Alibaba, Wan 2.1 will come in four different versions named “T2V-1.3B,” “T2V-14B,” “I2V-14B-720P,” and “I2V-14B-480P.” The most powerful version is designed to process 14 billion parameters. However, this AI model has a significant limitation; even with the most advanced version, the highest resolution you can achieve is 720p.
Here’s a video showcasing what Wan 2.1 can do
As shown in the video above, Alibaba’s new AI model can produce incredibly realistic videos. The model has reached a level where distinguishing between real and fake videos is nearly impossible. This is not surprising, as the Wan 2.1 model has been rated as the best model in this field according to VBench, a benchmark for video-generating AI models.
If you’re eager to try out Alibaba’s new AI model, you can access GitHub through this link and Hugging Face through this link.
