Bytedance
Developed by Bytedance, who is a newcomer to the market for large language models (LLM), its homonymous chatbot is the application of the most popular consumers in China.
The last model of LLM of the company, Duobao-1.5-Pro, was launched in January and surpasses Chatgpt’s GPT-4 in some metrics.
Bytedance is among the leaders in the development of LLM of low cost, and its offer is the most affordable available in the market, even more economical than Depseek, according to a Barclays report.
Duobao-1.5 has a price that represents only 3% to 4% of the price of GPT-4.
The advantage of costs by costs is attributed to the use of the frame “mixture of experts” (MOE), common in several AI models in China, including those of Deepseek.
Qwen
Developed by the Electronic Commerce giant Alibaba, Qwen is the company LLM and chatbot of the company. It was released in March 2023 and has been qualified among the best at the global reference points.
In January, Alibaba launched Qwen 2.5-Max and said that its functionality exceeded that of Deepseek-V3, highly acclaimed. Firm defender of the development of AI open, Alibaba announced in September that many of its AI models would be open source.
Ernie
The Chinese searches giant Baidu was one of the first in China to launch an equivalent of Chatgpt, Ernie Bot. Baidu states that his latest model, Ernie 4.0, equals GPT-4 capabilities.
On February 14, Baidu announced a plan to make its LLM next generation, Ernie 4.5, open source from June 30, an important change in its strategy as competition increases.
Hunyuan
Tencent launched its flagship model in September and launched its chatbot, Yuanbao, in May 2024. The huge user base of the company, acquired through its popular consumer applications such as Wechat, gives Tencent a significant advantage in the Impulse of applications for consumers in the future.
Tencent also offers open source models and advocates open source development. Another advantage of Hunyuan is its focus on multimodality, which includes text, image and video generation capabilities.
GLM
Zhipu AI developed GLM and states that his last LLM, GLM4, surpasses GPT-4 in some metrics. Initially originated in a laboratory of the University of Tsinghua, Zhipu has Alibaba, Tencent and the Zhongguancun Science City Innovation Development state among its sponsors.
In January, the US Department of Commerce included Zhipu in a list of entities subject to export controls.
Kimi
Developed by MoNshot, backed by companies such as Alibaba, Tencent and Hongshan. Its Kimi chatbot is known for its long -context processing technology, and the company states that the model admits two million entrance tokens, or words units, by consultation.
On January 20, only two days before Depseek launched the R1 model, Mososhot presented its own reasoning model, Kimi 1.5.
Minimax
Minimax is a pioneer among Chinese companies in the investigation of Marco Moe and launched its most advanced model, Minimax-01, open source in January. Minimax was founded in December 2021, long before the rise of the AI generated by the launch of OpenAi chatpt. It is backed by Alibaba and other prominent investors, such as Hongshan, Gaorong Capital and IDG.
In addition to LLMS, Minimax has gained recognition for its social applications for consumers, such as Takie and Xingye.
01.AI
Similar to Depseek, 01.AI has adopted an open source approach and has been a pioneer in the MOE Marco. The company states that its models train with less resources and have a cost advantage, reaching the lowest level of inference of LLM of the industry, 1/40 of the GPT-4 list price.
Its founder Kai-Fu Lee, former Chinese chief, said recently in a written interview with Reuters that 01. AI has formed a joint association with Alibaba to create a laboratory focused on continuing the development of LLM technologies. The company is backed by companies such as Alibaba and Sining Ventures.
Baichuan
Baichuan was founded in April 2023 by a team of former executives of the SOGOU search engine, including Founder and CEO Wang XiaoChuan, who was previously CEO of SOGOU.
The company won recognition in June 2023 by becoming the first in China to launch an open source model.
Source: Ambito
I am a 24-year-old writer and journalist who has been working in the news industry for the past two years. I write primarily about market news, so if you’re looking for insights into what’s going on in the stock market or economic indicators, you’ve come to the right place. I also dabble in writing articles on lifestyle trends and pop culture news.
