Chinese tech veteran, Baidu, is creating its Erney Artificial Intelligence (AI) model open-source on Monday. The company announced the move in February, and has already made its chatbot free to use. It is seen as a watershed moment in China’s AI Space, as this is the first time a leading AI player has decided to release his ownership technology to the open community. According to reports, experts believe that Baidu’s decision may have a similar impact on the global AI market as the Deepsek-R1 had emerged for the first time.
Baidu for open-source eri in a gradual way
On 14 February, Baidu confirmed that its major big language model (LLM) ERNIE Will be made available for open community. The WeChat post mentioned that the company will open its “next generation AI model”, which many believe that Erni is 5. In particular, in March, The company released the Erni 4.5 Foundation ModelDubbed Erni X1 with its first argument model.
While the company is turning its strategy into open-sources, it is not currently known whether the AI models will be completely open or partially. East means when a company or firm model models release details about weight, training cookbook and architecture. It not only enables developers to use models independently, but also modifies, replication and construction on them.
However, in the latter case, only model weight is open, so the company still keeps other details private. With this, developers can only use models and fine-tune, but cannot repeat it. An example of a fully open-source AI model is Mistral Nemo 12B, while a Meta Lama with only open weight is 3.
Experts believe that the move can have a positive effect on making AI democratic worldwide. Sean Rain, Associate Professor of Computer Science at the University of Southern California and AI researcher of Samsung of the year, told CNBC, “It’s not just a story of China. Every time a major laboratory opens a powerful model, it increases the bar for the whole industry.”
Interestingly, the last time a Chinese Open-SOS AI model was released (Deepsek-R1), it completed the stock prices of major AI players, and was, and was A net negative effect On most US-based companies.