TechFlow news: AIatMeta, the official account of Meta, announced that Meta has officially released two open-source AI language models under Llama 3, featuring 8 billion and 70 billion parameters respectively. The models achieve new capabilities such as improved reasoning performance and set new state-of-the-art benchmarks for specific tasks.
It is reported that in the coming months, the team plans to roll out additional features, longer context windows, more model sizes, enhanced performance, and a Llama 3 research paper, enabling the community to learn from the advancements in Llama 3.
Add to Favorites
Share to Social Media
