Alibaba Expands AI Offerings with Open-Source Models and Text-to-Video Generation

Alibaba Expands AI Offerings in Bid for Open Source | PYMNTS.com

Alibaba's release of over 100 open-source artificial intelligence models and text-to-video technology signals its growing focus on AI for digital commerce. The Chinese tech company's expansion in generative AI aims to provide businesses with new tools for customer engagement and content creation in online retail and marketing. The move could democratize access to advanced AI tools, particularly benefiting small- to medium-sized businesses (SMBs). "Large tech incumbents, whether Alibaba or Meta, are playing the age-old game of commoditizing your complement," Mike Conover, CEO of AI company Brightwave, told PYMNTS. "By creating a robust development ecosystem around technologies that enhance their core offerings, such as product marketplaces or messaging, these firms increase their differentiated value and talent brand in a competitive hiring environment." The release could lower barriers for businesses, especially SMBs, to adopt advanced AI tools in eCommerce and digital marketing. This democratization of AI access may lead to more personalized customer experiences, innovative product showcases and efficient operations across the global retail landscape, potentially reshaping how companies compete and engage consumers in the digital marketplace. Alibaba's open-source AI strategy allows global users to create generative AI applications without the costly process of training their systems. This approach has led to 40 million downloads of their models since the launch of Tongyi Qianwen (Qwen) last year, according to a company blog post. Alongside open-source offerings, Alibaba upgraded its proprietary Qwen-Max model, claiming similar performance as competitors like Meta's Llama and OpenAI's GPT4 in reasoning and language comprehension, per the post. The company also introduced a text-to-video tool like OpenAI's Sora, expanding its AI capabilities. The multi-pronged approach aims to increase Alibaba's AI user base and enhance its position in the global AI market. "Open-source AI large models have the potential to significantly impact global commerce and trade by democratizing access to advanced AI tools, reducing barriers for businesses of all sizes to leverage AI," Pundi X co-CEO Peko Wan told PYMNTS. "[SMBs] will benefit from reduced costs, allowing them to leverage AI for innovation and competitiveness." Wan also pointed out potential benefits for Alibaba. "Alibaba can attract a broader user base, encouraging businesses to adopt their cloud infrastructure for hosting and scaling AI solutions, thereby driving increased demand for their paid cloud services like storage, computing power and data management," Wan said. The tech industry has seen a trend of open-source AI. These publicly accessible tools offer a range of capabilities, from solving complex mathematical problems to writing code and translating between 29 languages, the post said. Industry experts point out that this approach, also adopted by other tech giants, reshapes the AI landscape. Meta's Llama, for instance, has been widely used for natural language processing tasks, while Google's BERT has become a cornerstone for many text analysis applications. By making these AI models freely available, companies enable a broader range of users to develop AI applications. A small eCommerce startup, for example, could use these models to create a chatbot for customer service or an automated product recommendation system. Similarly, researchers in fields like climate science or genomics could use these tools for data analysis without needing extensive AI expertise or resources. The democratization of AI technology is expected to accelerate innovation across various sectors. In healthcare, open-source models could be adapted for tasks like medical image analysis or drug discovery. In finance, they might be used to enhance fraud detection systems or automate risk assessments. As AI increasingly influences eCommerce, Alibaba's move may prompt competitors to reconsider their AI strategies and could shift how companies approach digital sales and advertising globally. The long-term impact on international commerce could be significant, with Wan suggesting it may "foster greater collaboration across borders, breaking down traditional barriers in international trade and creating more dynamic, interconnected global markets." Thanks to cheaper open-source models, smaller businesses and startups could access sophisticated AI capabilities, enabling them to compete more effectively with larger players in the eCommerce space. Swapnil Sawant, a software developer at Phoenix Bioinformatics, told PYMNTS: "With basic AI capabilities now openly accessible, competition will shift toward how effectively eCommerce companies can integrate and customize these models, as well as the development of more advanced, specialized AI applications with better UX integrated with AI enhancing online shopping further. The global adoption of these models may spark discussions about data privacy and digital sovereignty, potentially influencing international trade policies and regulations."

China's Alibaba launches over 100 new open-source AI models, releases text-to-video generation tool

AI models are trained on huge amounts of data. Alibaba says its models have the abiltiy to understand prompts and generate texts and images. Open-source means that anyone -- including researchers, academics and companies -- around the world can use the models to create their own generative AI apps without needing to train their own systems, saving time and expense. By open sourcing the models, Alibaba hopes more users will use its AI. The Chinese e-commerce giant first launched its Tongyi Qianwen, or Qwen, model last year. Since then, it has released improved versions and says that, to date, its open source models have been downloaded 40 million times. The company also said that it upgraded its proprietary flagship model called Qwen-Max, which is not open-source. Instead, Alibaba sells its capabilites through its cloud computing products to businesses. Alibaba said that Qwen Max 2.5-Max surpassed rivals such as Meta's Llama and OpenAI's GPT4 in several areas inclduing reasoning and language comprehension. Alibaba also launched a new text-to-video tool based on its AI models. This allows users to input a prompt and the AI will create a video based on it. This is similar to OpenAI's Sora. "Alibaba Cloud is investing, with unprecedented intensity, in the research and development of AI technology and the building of its global infrastructure," Eddie Wu, CEO of Alibaba, said in a statement.

Alibaba accelerates AI push by releasing new open-source models, text-to-video

BEIJING - Chinese technology company Alibaba released on Thursday new open-source artificial intelligence models and text-to-video AI technology, intensifying its efforts to compete in the booming area of generative AI. The open-source models, numbering more than 100, come from Alibaba's Qwen 2.5 family, its latest foundational large language model released in May. Chinese technology companies, like their U.S. counterparts, have been investing heavily in generative AI, with firms racing to develop robust product portfolios and diversified offerings. While competitors such as Baidu and OpenAI have primarily adopted closed-source approaches, Alibaba has embraced a hybrid model, investing in both proprietary and open-source development to broaden its AI product range. Its new models range from 0.5 to 72 billion parameters - the variables that determine an AI model's capability and performance - in size, offering proficiency in mathematics, coding, and support for over 29 languages, Alibaba said in a statement. The models aim to cater to a wide array of AI applications across various sectors including automotive, gaming, and scientific research. Alibaba on Thursday also unveiled a new text-to-video model as part of its Tongyi Wanxiang image generation family, joining a growing number of Chinese tech firms entering this emerging market. The move puts Alibaba in direct competition with global players such as OpenAI, which has also shown interest in text-to-video technology. In August, ByteDance, owner of TikTok, launched its text-to-video app Jimeng AI on Apple's App Store for Chinese users. (Reporting by Liam Mo and Casey Hall; Editing by Muralikumar Anantharaman)

Alibaba accelerates AI push by releasing new open-source models, text-to-video

BEIJING, Sept 19 (Reuters) - Chinese technology company Alibaba (9988.HK), opens new tab released on Thursday new open-source artificial intelligence models and text-to-video AI technology, intensifying its efforts to compete in the booming area of generative AI. The open-source models, numbering more than 100, come from Alibaba's Qwen 2.5 family, its latest foundational large language model released in May. Chinese technology companies, like their U.S. counterparts, have been investing heavily in generative AI, with firms racing to develop robust product portfolios and diversified offerings. Advertisement · Scroll to continue While competitors such as Baidu (9888.HK), opens new tab and OpenAI have primarily adopted closed-source approaches, Alibaba has embraced a hybrid model, investing in both proprietary and open-source development to broaden its AI product range. Its new models range from 0.5 to 72 billion parameters - the variables that determine an AI model's capability and performance - in size, offering proficiency in mathematics, coding, and support for over 29 languages, Alibaba said in a statement. Advertisement · Scroll to continue The models aim to cater to a wide array of AI applications across various sectors including automotive, gaming, and scientific research. Alibaba on Thursday also unveiled a new text-to-video model as part of its Tongyi Wanxiang image generation family, joining a growing number of Chinese tech firms entering this emerging market. The move puts Alibaba in direct competition with global players such as OpenAI, which has also shown interest in text-to-video technology. In August, ByteDance, owner of TikTok, launched its text-to-video app Jimeng AI on Apple's (AAPL.O), opens new tab App Store for Chinese users. Reporting by Liam Mo and Casey Hall; Editing by Muralikumar Anantharaman Our Standards: The Thomson Reuters Trust Principles., opens new tab

Alibaba accelerates AI push by releasing new open-source models, text-to-video

Chinese technology companies, like their U.S. counterparts, have been investing heavily in generative AI, with firms racing to develop robust product portfolios and diversified offerings. While competitors such as Baidu and OpenAI have primarily adopted closed-source approaches, Alibaba has embraced a hybrid model, investing in both proprietary and open-source development to broaden its AI product range. Its new models range from 0.5 to 72 billion parameters - the variables that determine an AI model's capability and performance - in size, offering proficiency in mathematics, coding, and support for over 29 languages, Alibaba said in a statement. The models aim to cater to a wide array of AI applications across various sectors including automotive, gaming, and scientific research. Alibaba on Thursday also unveiled a new text-to-video model as part of its Tongyi Wanxiang image generation family, joining a growing number of Chinese tech firms entering this emerging market. The move puts Alibaba in direct competition with global players such as OpenAI, which has also shown interest in text-to-video technology. In August, ByteDance, owner of TikTok, launched its text-to-video app Jimeng AI on Apple's App Store for Chinese users. (Reporting by Liam Mo and Casey Hall; Editing by Muralikumar Anantharaman)

Alibaba accelerates AI push by releasing new open-source models, text-to-video

Alibaba Group sign is seen at the World Artificial Intelligence Conference in Shanghai, China, on July 6, 2023. -- Reuters file Chinese technology company Alibaba released on Thursday new open-source artificial intelligence models and text-to-video AI technology, intensifying its efforts to compete in the booming area of generative AI. The open-source models, numbering more than 100, come from Alibaba's Qwen 2.5 family, its latest foundational large language model released in May. Chinese technology companies, like their US counterparts, have been investing heavily in generative AI, with firms racing to develop robust product portfolios and diversified offerings. While competitors such as Baidu and OpenAI have primarily adopted closed-source approaches, Alibaba has embraced a hybrid model, investing in both proprietary and open-source development to broaden its AI product range. In August, ByteDance, owner of TikTok, launched its text-to-video app Jimeng AI on Apple's App Store for Chinese users.

Alibaba releases over 100 open-source AI models open source and new text-to-video generator - SiliconANGLE

Alibaba releases over 100 open-source AI models open source and new text-to-video generator Alibaba Cloud, the cloud computing arm of China's Alibaba Group Ltd., today announced the release of over 100 new artificial intelligence large language models open source as part of the Qwen 2.5 family of models. Revealed at the company's Apsara Conference, the new model series follows the release of the company's foundation model Tongyi Qianwen, or Qwen, last year. Since then, the Qwen models have been downloaded over 40 million times across platforms such as Hugging Face and Modelscope. The new models range from sizes as small as 0.5 billion parameters to as large as 72 billion parameters. In an LLM parameters define the behavior of an AI model and what it uses to make predictions about its skills such as mathematics, coding or expert knowledge. Smaller, more lightweight models can be trained quickly using far less processing power on more focused training sets and excel at simpler tasks. In contrast, larger models need heavy processing power and longer training times and generally perform better on complex tasks requiring deep language understanding. Within the Qwen 2.5 family, Alibaba included models that provide generalized language knowledge for summarization, question and answer and other standard LLM responses. The company also released models for coding and mathematics skill sets that the company said have undergone substantial enhancements compared to the previous generations. "Alibaba Cloud is investing, with unprecedented intensity, in the research and development of AI technology and the building of its global infrastructure," said Eddie Wu, chairman and chief executive of Alibaba Cloud Intelligence. "We aim to establish an AI infrastructure of the future to serve our global customers and unlock their business potential." In addition to the language models, the company also unveiled updates to its multimodal AI models including a new text-to-video model as part of its image generator, Tongyi Wanxiang large model family. The new model can generate high-quality videos based on Chinese and English prompts and transform static images into high-quality videos in a wide variety of styles. The videos that the model can produce include realistic scenes and 3D animation. Under the hood, the model features advanced diffusion transformer architecture to enhance video reconstruction quality.

Alibaba Makes Over 100 Large Language Models Open Source

Alibaba is making over 100 of its large language models available to the global open-source community, as part of its growing push into artificial intelligence. Jingren Zhou, chief technology officer of Alibaba Cloud Intelligence described the initiative as a "significant milestone" and the company's most expansive open-source initiative to date. The new open source models, under Alibaba Cloud's Qwen 2.5 series portfolio of large language models, includes different modalities such as language, audio, and vision, according to the company. The announcement was made during Alibaba Cloud's annual three-day event in Hangzhou, China, where it invited well-known AI start-up founders and scientists, including Li Feifei, a Chinese-American computer scientist and former board director at Twitter. Since its debut in April 2023, Alibaba's Qwen models have exceeded 40 million downloads across open-source platforms such as Hugging Face and ModelScope, Alibaba said.