AMD benchmarks 512 MI355X GPU cluster in MLPerf 6.1 while NVIDIA retains strong position with Blackwell & first Vera Rubin submit, Intel Arc Pro & Xeon submitted too

MLPerf Inference v6.1 results are out. AMD tested a 512-GPU Instinct MI355X cluster for the first time, Nvidia previewed Vera Rubin NVL72 with up to 3.7x gain over GB300, and Intel expanded participation with Xeon 6 and Arc Pro B70.

• AMD: 512 MI355X delivered 5.75M tokens/s on GPT-OSS-120B and 2.90M on DeepSeek-R1
• Nvidia: 72-GPU Vera Rubin up to 95% faster than 72-GPU GB300, 36-GPU VR200 faster than 72-GPU GB200
• Intel expanded Xeon 6 participation from 2 to 5 SKUs, CPU results grew from 24 to 35
• Arc Pro B70 on 4 GPUs: +36% Server and +27% Offline on gpt-oss-120B vs v6.0