Alibaba Group, the Chinese tech and e-commerce giant, revealed on August 3 that its cloud computing division has introduced two open-source artificial intelligence (AI) models, as stated in a press release.
These models, called Qwen-7B and Qwen-7B-Chat, are considered large language models (LLMs), with each having 7 billion parameters. These newly unveiled models are scaled-down versions of the Uniform Qiawen model that Alibaba launched in April.
The primary objective of these models is to integrate artificial intelligence capabilities into the operations of small and medium-sized businesses. Qwen-7B and Qwen-7B-Chat offer a range of features that make them appealing to enterprises. One significant aspect is that they grant "free access to code, model weights, and documentation by academics, researchers, and commercial organizations around the world," according to the company's statement.
This move marks Alibaba's initial contribution of an open-source large language model in the AI domain, representing a notable development for the Chinese technology industry. Nonetheless, businesses with over 100 million monthly active users will require a license to utilize these models s. As of On August 1, Alibaba also announced a vector engine update to its AnalyticDB data warehouse service, which empowers enterprise clients to quickly develop custom generative AI applications.
This announcement comes in the wake of Meta and Microsoft unveiling their open-source Llama 2 model on July 16. Meta's Llama 2 is trained on a greater proportion of public data, handling twice as many contexts as its predecessor. While it's open source, a license from companies with more than 700 million monthly users is necessary to use the largest version, Llama 2. On July 26, Alibaba introduced "China's first Llama 2 full-series training and deployment solution," having previously developed AI software and tools for businesses to implement Llama 2 solutions.
