Search
MiniMax Unveils M1: The Pioneering Open-Source Large-Scale Hybrid Architecture Inference Model
1 week ago /
Read about 0 minute
Author：小编
MiniMax XiYu Technology has announced a groundbreaking series of updates spanning five consecutive days, kicking off with the open-sourcing of the revolutionary MiniMax-M1 inference model. As the world's first open-source large-scale hybrid architecture inference model, MiniMax-M1 boasts an industry-leading capability, supporting an unprecedented input context of up to 1 million tokens and generating outputs of up to 80,000 tokens. Remarkably, its training cost is a mere tenth of comparable models, setting a new benchmark in efficiency and accessibility.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
