Open source AI models

Hunyuan Video is a state-of-the-art open source AI video generation model developed by Tencent with 13 billion parameters. It is capable of generating high-quality videos based on text descriptions with excellent motion quality and visual fidelity. It supports resolutions up to 720p x 1280p and is particularly good at generating motion diversity and high alignment of text and video. Hunyuan Video Features Powerful AI Video Generation Capabilities:Hunyuan Video is an open source AI video generation model based on 13 billion parameters, capable of generating videos based on...

Recently, Cohere announced the launch of two new open-source AI models that aim to close the language gap in the base model through its Aya project. The two new models, called Aya Expanse8B and 35B, are now available on Hugging Face. The launch of these two models has resulted in significant AI performance improvements in 23 languages. The 8B parametric model makes breakthroughs easier for researchers worldwide, while the 32B parametric model provides industry-leading multilingual capabilities, Cohere said in his blog post. Aya...

Recently, LG AI Research announced the launch of Exaone3.0, South Korea's first open source artificial intelligence model. The model has 7.8 billion parameters and can perform well in Korean and English tasks. It aims to accelerate AI research and build a strong AI ecosystem for South Korea. LG has always been famous for its consumer electronics, and this strategic transformation shows that it hopes to gain a foothold in the field of AI innovation. By open sourcing Exaone3.0, LG not only demonstrates its technological strength, but also may open up new revenue sources for cloud computing and AI services…

Google released a press release yesterday, releasing the Gemma 2 large language model to researchers and developers around the world. There are two sizes: 9 billion parameters (9B) and 27 billion parameters (27B). Compared with the first generation, the Gemma 2 large language model has higher reasoning performance, higher efficiency, and has made significant progress in security. Google stated in the press release that the performance of the Gemma 2-27B model is comparable to mainstream models of twice the size, and only requires an NVIDIA H100 ensor Core GPU or TPU main...

In the latest research by USTC and other institutions, scientists have released an important achievement, namely SciGuard and SciMT-Safety. The goal of this innovative approach is to protect AI for Science models to prevent them from being improperly used in biology, chemistry, medicine and other fields. To this end, the research team also established SciMT-Safety, the first benchmark focused on safety in the field of chemical science. Paper address: https://arxiv.org/pdf/2312.06632.pdf The research team revealed the existing open…