16 × NVIDIA B300-SXM-270GB

Tested system

Processor: Intel(R) Xeon(R) 6776P
Graphics: NVIDIA B300-SXM-270GB
Accelerator count: 16
Nodes: 2

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

MLPerf Inference / deepseek-r1Offline · v6.1 · Tokens/s

16 × NVIDIA B300-SXM-270GB

136,854^* Tokens/s

Accuracy target / result: exact_match: 81.24430264357338 TOKENS_PER_SAMPLE: 3826.166362807657
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 2
Total accelerators: 16
Software stack: TensorRT 10.14, CUDA 13.2
Operating system: RHEL 9.4

MLPerf Inference / deepseek-r1Server · v6.1 · Tokens/s

16 × NVIDIA B300-SXM-270GB

80,536.90^* Tokens/s

Accuracy target / result: exact_match: 80.92525068368278 TOKENS_PER_SAMPLE: 3807.0186873290795
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 2
Total accelerators: 16
Software stack: TensorRT 10.14, CUDA 13.2
Operating system: RHEL 9.4

MLPerf Inference / gpt-oss-120bOffline · v6.1 · Tokens/s

16 × NVIDIA B300-SXM-270GB

114,569^* Tokens/s

Accuracy target / result: exact_match: 83.235
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 1
Total accelerators: 8
Software stack: TensorRT 10.14, CUDA 13.2
Operating system: RHEL 9.4

MLPerf Inference / gpt-oss-120bServer · v6.1 · Tokens/s

16 × NVIDIA B300-SXM-270GB

113,685^* Tokens/s

Accuracy target / result: exact_match: 83.448
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 1
Total accelerators: 8
Software stack: TensorRT 10.14, CUDA 13.2
Operating system: RHEL 9.4

Processing & platform

Processor: Intel(R) Xeon(R) 6776P
Graphics: NVIDIA B300-SXM-270GB

Connectivity & software

Operating system: RHEL 9.4

System configuration

Reported host core count: 64
Processors per node: 2
Accelerator count: 16
Nodes: 2
Software stack: TensorRT 10.14, CUDA 13.2
Submission division: closed

* Provisional data