It is 671 billion parameters, not 671 亿.

来源: car_accident 于 2025-01-30 20:17:11 [] [旧帖] [给我悄悄话] 阅读数 : (371 bytes)

 "Combined with the software optimizations available in the NVIDIA NIM microservice, a single server with eight H200 GPUs connected using NVLink and NVLink Switch can run the full, 671-billion-parameter DeepSeek-R1 model at up to 3,872 tokens per second".

请您先登陆，再发跟帖！