The arrival of the 86B model represents a significant change in the field of large language models. Its relatively smaller scale compared to behemoths like GPT-4 shouldn't necessarily suggest weakness, 86blegal.com