Search AI Products and News

Zhihu AI has officially released CogView4, its latest open-source text-to-image model. Boasting 600 million parameters, CogView4 notably supports both Chinese and English prompts and text-to-image generation, making it the first open-source model capable of generating Chinese characters within images. Its key feature is its support for bilingual prompts, excelling at understanding and following complex Chinese instructions, a boon for Chinese content creators. As the first open-source text-to-image model to generate Chinese characters in images, it fills a significant gap in the open-source landscape.

The Ministry of Education recently announced that the expert committee of the inter-ministerial joint conference on foreign language Chinese translation standards has reviewed and approved the 16th batch of 18 groups of recommended Chinese translations for foreign terms. These translations primarily relate to the fields of artificial intelligence and information technology. This initiative aims to standardize the new terms that frequently appear in the AI field, providing the public with clear official explanations to eliminate confusion.

["The Southern University of Science and Technology and the IDEA Research Institute CCNL Center have released SUS-Chat-34B, a bilingual model with a parameter scale of 34 billion.", "This model excels in various Chinese and English tasks, surpassing other models of the same parameter scale.", "SUS-Chat-34B is fine-tuned from the pre-trained model 01-ai/Yi-34B using millions of high-quality, multilingual instruction data.", "The model has a large-scale complex instruction following dataset and strong capabilities." ]

ZhiYuan Research Institute has unveiled the new open-source bilingual model Wudao・Tianying 34 Billion Aquila2-34B, which excels in reasoning, generalization, and more. The institute has also released a comprehensive open-source toolkit to promote collaborative innovation in large model research. Aquila2-34B surpasses other open-source foundational models in overall capabilities, with the ZhiYuan team developing the NLPE method to enhance the model's extension capabilities.