Search
On October 15th, Ant Bailing made a significant announcement regarding the official open-sourcing of its latest-generation large-scale model, Ling-3.1-flash. This cutting-edge model boasts an impressive total of approximately 560 billion parameters, with around 25 billion being activated for each token. It also supports a maximum context window of 1 million, rendering it highly suitable for handling lengthy documents, code snippets, and task histories. The primary objective behind this open-sourcing initiative is to empower teams to evaluate the model's capabilities and inference efficiency within their specific task contexts and computing environments, aligning with the best practices in the tech industry where open collaboration often leads to rapid innovation and improvement.
-
C114 Communication Network -
Communication Home
