Media365
According to the model’s technical report, LongCat-Flash-Chat features 560 billion total parameters – the variables used to define and manipulate the input and output of models – and the so-called Mixture-of-Experts (MoE) architecture. MoE architecture divides the model into separate sub-networks, or “experts”, that specialise in a subset of the input data to jointly perform a task.
Meituan’s foray into open-source models reflects its efforts to establish a viable AI business since acquiring Light Year, which was established by a co-founder of the food delivery services giant.
It also showed how Chinese AI companies are continuing to narrow the gap with their US peers through the open-source approach, which makes the source code of AI models available for third-party developers to use, modify and distribute.
