Content by Mark Gitau and Azin Heidarshenas (1)
Mark Gitau and Azin Heidarshenas summarize Azure’s MLPerf Inference v6.1 submissions for DeepSeek-R1 on NVIDIA GB300/GB200 NVL72, highlighting throughput and latency-focused results across Interactive, Server, and Offline scenarios at 1-rack and 4-rack scale.
End of content