11.2 C
Vancouver

Alibaba Open-Sources AI Model for Generating Realistic Videos and Images

Published:

BEIJING, Feb 26 Alibaba has announced that its latest generative AI model, Wan 2.1, is now open source, allowing users worldwide to access and modify the technology for free. The move is expected to drive greater adoption of Alibaba’s AI solutions and intensify competition in the rapidly evolving artificial intelligence space.

The Chinese tech giant has released four variants of Wan 2.1—T2V-1.3B, T2V-14B, I2V-14B-720P, and I2V-14B-480P—designed to generate highly realistic images and videos from both text and image inputs. The “14B” designation in certain models indicates they can process 14 billion parameters, enhancing their ability to deliver more precise and high-quality outputs.

Users can download the models from Alibaba Cloud’s ModelScope and Hugging Face, platforms that provide access to open-source AI models for academic, research, and commercial applications.

Growing Open-Source AI Momentum

Alibaba’s decision to open-source Wan 2.1 follows a broader trend in the AI industry. Last month, Chinese startup DeepSeek made waves by releasing its R1 reasoning model for free, sparking increased interest in accessible AI technologies. DeepSeek has also committed to expanding its open-source efforts, with plans to release additional code repositories behind its services.

Alibaba has been a key player in the generative AI space since the debut of OpenAI’s ChatGPT. The company’s AI technology is poised to power artificial intelligence features for Apple iPhones sold in China, where regulatory constraints prevent Apple from using its standard AI models. This partnership underscores Alibaba’s growing role in shaping AI applications within the Chinese market.

Advancements in AI and Cloud Investment

Since introducing the latest iteration of its video- and image-generating AI model in January, Alibaba has emphasized its technological superiority. The company has touted Wan 2.1’s top ranking on VBench, a benchmarking platform for video generative models, where it leads in key functionalities such as multi-object interaction.

In addition to its AI advancements, Alibaba has revealed ambitious investment plans. The company will allocate at least 380 billion yuan ($52 billion) over the next three years to enhance its cloud computing and AI infrastructure, reinforcing its commitment to maintaining a competitive edge in the sector.

Alibaba also previewed QwQ-Max, a reasoning model it plans to make open source upon its full release, further strengthening its position in the AI space.

With increasing momentum in open-source AI, Alibaba’s latest move signals a shift towards greater accessibility and innovation in artificial intelligence development.

This article originally appeared in The New Digital’s AI and Tech News website.

Related articles

spot_img

Recent articles

spot_img