Alibaba has officially released Wan 2.1, its advanced image and video-generating AI model, as an open-source platform. This move is set to fuel AI innovation and intensify competition with OpenAI’s Sora.
The Chinese tech giant unveiled four versions of Wan 2.1—T2V-1.3B, T2V-14B, I2V-14B-720P, and I2V-14B-480P—each designed to enhance the precision of AI-generated images and videos.
Key Features of Wan 2.1
- Powerful AI Processing: The 14B models leverage 14 billion parameters, enabling high-quality text-to-video, image-to-video, video editing, text-to-image, and video-to-audio capabilities.
- Multi-Language Support: The T2V-14B model uniquely generates videos with both Chinese and English text.
- Optimized for Consumer Hardware: The T2V-1.3B model can generate a five-second 480P video in just four minutes, requiring only 8.19GB of VRAM on an RTX 4090 GPU.
Alibaba’s Strategic AI Expansion
This open-source release aligns with a broader push for AI accessibility and follows in the footsteps of DeepSeek, another Chinese AI firm. Alibaba initially introduced its image and video generation AI under the name Wanx in January before rebranding it to Wan.
Additionally, Alibaba has previewed QwQ-Max, a powerful reasoning model, with plans to open-source it upon launch. The company has also committed a massive ¥380 billion ($52 billion) investment in cloud computing and AI infrastructure over the next three years.
By making Wan 2.1 publicly available, Alibaba is positioning itself as a global AI leader, setting the stage for intense competition in the rapidly evolving AI video and image generation space.
