Supermicro Systems with the NVIDIA B200 Outperform the Previous Generation by 3X

Super Micro Computer, Inc. (SMCI), a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge

Leading industry-standard performance in select MLPerf V5.0 Inference benchmarks is announced today by Super Micro Computer, Inc. (Nasdaq: SMCI), a premier provider of complete IT solutions for AI/ML, HPC, cloud, storage, and 5G/edge.

In several V5.0 Inference benchmarks, Supermicro systems achieved the best performance. The company's 4U liquid-cooled and 10U air-cooled systems, which used the NVIDIA HGX B200 8-GPU, outperformed H200 8-GPU systems by more than 3 times in the generation of tokens per second (Token/s) for the Llama2-70B and Llama3.1-405B benchmarks.

"We are pleased to see the first new benchmarks from MLCommons in 2025, with Supermicro remaining a leader in the AI industry," said Charles Liang, president and CEO of Supermicro. "Our building block architecture allows us to be first to market with a diverse range of systems optimized for various workloads. We continue to collaborate closely with NVIDIA to fine-tune our systems and secure a leadership position in AI workloads."

Learn more about the new MLPerf v5.0 Inference benchmarks at: https://mlcommons.org/benchmarks/inference-datacenter/

Supermicro is the only system vendor to publish record MLPerf inference performance (on select benchmarks) for both the air-cooled and liquid-cooled NVIDIA HGX B200 8-GPU systems. Both air-cooled and liquid-cooled systems were operational before the MLCommons benchmark start date. Supermicro engineers optimized the systems and software to showcase the impressive performance.

Within the operating margin, the Supermicro air-cooled B200 system exhibited the same level of performance as the liquid-cooled B200 system. Supermicro has been delivering these systems to customers while we conducted the benchmarks.

MLCommons emphasizes that all results be reproducible, that the products are available and that the results can be audited by other MLCommons members. Supermicro engineers optimized the systems and software, as allowed by the MLCommons rules.

The SYS-421GE-NBRT-LCC (8x NVIDIA B200-SXM-180GB) and SYS-A21GE-NBRT (8x NVIDIA B200-SXM-180GB) showed performance leadership running the Mixtral 8x7B Inference, Mixture of Experts benchmarks with 129,000 tokens/second. The Supermicro air-cooled and liquid-cooled NVIDIA B200 based system delivered over 1,000 tokens/second inference for the large Llama3.1-405b model, whereas the previous generations of GPU systems have much smaller results. For smaller inferencing tasks, using the LLAMA2-70b benchmark, a Supermicro system with the NVIDIA B200 SXM-180GB installed shows the highest performance from a Tier 1 system supplier.

Specifically:

* The SYS-421GE-NBRT-LCC (8x NVIDIA B200-SXM-180GB) liquid-cooled system achieved the highest performance for the Mixtral 8x7B, Mixture of Experts benchmark, delivering 129,000 tokens/s. It also achieved the highest performance for the Llama3.1-405b benchmark with 1,330 tokens/s.

* The SYS-A21GE-NBRT (8x NVIDIA B200-SXM-180GB) air-cooled system achieved the highest performance for the Mixtral 8x7B, Mixture of Experts benchmark, delivering 129,000 tokens/s. It achieved the second-highest performance for the Llama3.1-405b benchmark with 1,009 tokens/s.

* The SYS-110R3-L (8x NVIDIA A200-SXM-80GB) system achieved the highest performance for the BERT Large benchmark.

* The SYS-110R3-L (8x NVIDIA A200-SXM-80GB) and SYS-A21GE-NBRT (8x NVIDIA B200-SXM-180GB) systems achieved the highest performance for the DeepLabV3 and REsNet-50 benchmarks.

* The SYS-110R3-L (8x NVIDIA A200-SXM-80GB) system achieved the highest performance for the 3D object detection benchmark.

Supermicro offers a comprehensive AI portfolio with over 100 GPU-optimized systems, both air-cooled and liquid-cooled options