Alibaba Cloud Unveils Open-Sourced AI Models to Challenge Meta’s Llama 2

The move follows Meta’s release of its own open-sourced LLM, Llama 2, in July

Alibaba Cloud, the cloud computing arm of the Chinese tech giant Alibaba Group, has introduced two open-sourced artificial intelligence (AI) language models named Gwen-7B and Gwen-7B-Chat. With each model boasting up to 7 billion parameters, the company has become the first major Chinese enterprise to embrace open-source LLM. In an official press release on August 3, the company said the two open-sourced language models were built with up to 7 billion parameters each, making them potent contenders in the AI landscape.

The move follows Meta‘s release of its own open-sourced LLM, Llama 2, in July. While Alibaba’s Gwen models have 7 billion parameters each, Meta’s Llama 2, a commercial version of its open-sourced model Llama, has 70 billion parameters, making it one of the largest open-sourced LLMs globally. Last week, Alibaba Cloud partnered with Meta to make Llama accessible to Chinese businesses, enabling the companies to create customized AI applications.

Alibaba Cloud Makes New AI Models Available for Free

The Alibaba cloud unit had previously introduced its LLM named Tongyi Qianwen in April, offering various versions with different parameter counts. The newly open-sourced Gwen-7B and Gwen-7B-Chat are smaller-size iterations of Tongyi Qianwen, catering to small and medium businesses seeking to harness the power of AI.

The company aims to make the code, model weights, and documentation of the two models freely accessible to users, including academic researchers and commercial institutions worldwide. However, entities with over 100 million monthly active users will require a license from Alibaba to utilize the models. Similarly, Meta’s Llama 2 also demands a permit from the company for companies with over 700 million users.

The open-sourcing of these large language models is part of the effort to help democratize AI technology. By making these models available for free, more businesses and developers will be able to use them to develop new AI-powered applications.

LLM AI models can process massive amounts of data and generate content across various formats, including text, images, audio, and video. These tools serve as the backbone of AI chatbots like ChatGPT, driving innovation and enhancing user experiences across multiple domains.

Chinese Authorities Encourage AI Development in the Country

While Alibaba Cloud has not disclosed the higher-parameter versions of Tongyi Qianwen, this move signifies a significant stride in China’s bid to catch up with the United States in the AI domain. The Chinese government actively encourages local companies like Alibaba, Tencent, and Huawei to develop competitive and “controllable” AI models, challenging the supremacy of US-based AI technologies.

The open-sourcing of Alibaba Cloud’s LLM is expected to advance the technical capabilities and industrial applications of Chinese LLMs. Furthermore, the tools will simplify the model training and deployment process for enterprises, lowering the threshold for AI application and enabling the creation of tailor-made industry-specific LLMs with greater efficiency.

Meanwhile, aside from Alibaba and Meta, other tech giants like Google and Microsoft are also working on large language models for various applications, including text generation, language translation, and question-answering systems.

The global AI landscape is witnessing intense competition and rapid innovation, driving the growth and adoption of AI technologies in diverse industries, including crypto.

Artificial Intelligence, Business News, News, Technology News

Chimamanda U. Martha

Chimamanda is a crypto enthusiast and experienced writer focusing on the dynamic world of cryptocurrencies. She joined the industry in 2019 and has since developed an interest in the emerging economy. She combines her passion for blockchain technology with her love for travel and food, bringing a fresh and engaging perspective to her work.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *