Search
Moore Threads Achieves Swift Integration of MiniMax M2.5 Model
9 hour ago /
Read about 0 minute
Author：小编
On February 14, 2026, Moore Threads made a significant announcement: its premier AI training and inference GPU, the MTT S5000—a versatile, all-in-one solution—has swiftly accomplished Day-0 compatibility with MiniMax's latest large-scale model, the MiniMax M2.5. Engineered explicitly for extensive model training, inference tasks, and high-performance computing applications, the MTT S5000 boasts an impressive single-card AI computing capacity of up to 1000 TFLOPS. It is furnished with 80GB of memory and supports comprehensive precision computing, ranging from FP8 to FP64. This achievement underscores Moore Threads' formidable prowess in the realm of AI computing capabilities.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
