Chinese e-commerce giant Alibaba has launched a new AI model that it claims to have the ability to read human emotions. The model is called “R1-Omni,” and it is developed by Alibaba’s Tongyi Lab. By reading emotions, Alibaba’s R1-Omni aims to stay ahead of its rivals, such as OpenAI’s ChatGPT.
According to Bloomberg, the R1-Omni AI model is available as open source, and in demos, the AI model was able to understand human emotions from a video while also describing their clothes and surroundings. This new AI model from Alibaba enhances computer vision technology, building on a previous model called HumanOmni, which was created by the same lead researcher, Jiaxing Zhao.
Alibaba might want to bask in the success of DeepSeek, another Chinese AI company, which gained worldwide recognition after one of its AI models surpassed the likes of ChatGPT and shook the tech world. Alibaba is investing heavily in AI tools and applications, including its Qwen model, which it benchmarked against DeepSeek’s AI models.
Additionally, Alibaba has secured a partnership with Apple to bring AI features to iPhones in China. Now, Alibaba is stepping into OpenAI’s territory by offering R1-Omni, available for free on the Hugging Face platform.
OpenAI, meanwhile, has understood the assignment and the growing pace at which rival AI platforms are emerging. It launched its updated GPT-4.5 model earlier this year, which can detect subtle nuances in user prompts. However, the model isn’t available for free—it is accessible to premium subscribers who pay $200 a month.
Alibaba, on the other hand, wants to achieve its main goal of “artificial general intelligence.” The company’s CEO, Eddie Wu, made it clear to analysts in February that “artificial general intelligence” is the company’s topmost priority. And the latest developments in the field of AI are showcasing Alibaba’s push to establish itself as a major player in the global AI race
