4 × NVIDIA RTX PRO 6000 Blackwell Server Edition

Tested system

Processor: Intel(R) Xeon(R) 6747P
Graphics: NVIDIA RTX PRO 6000 Blackwell Server Edition
Accelerator count: 4
Nodes: 1

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

4 × NVIDIA RTX PRO 6000 Blackwell Server Edition

7,801.36^* Tokens/s

Accuracy target / result: exact_match: 83.969
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 48
Processors per node: 2
Nodes: 1
Total accelerators: 4
Software stack: TensorRT 10.14, CUDA 13.1
Operating system: Ubuntu 24.04

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

4 × NVIDIA RTX PRO 6000 Blackwell Server Edition

6,950.76^* Tokens/s

Accuracy target / result: exact_match: 83.637
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 48
Processors per node: 2
Nodes: 1
Total accelerators: 4
Software stack: TensorRT 10.14, CUDA 13.1
Operating system: Ubuntu 24.04

MLPerf Inference / llama2-70b-99.9Offline · v6.1 · Tokens/s

4 × NVIDIA RTX PRO 6000 Blackwell Server Edition

14,860.70^* Tokens/s

Accuracy target / result: ROUGE1: 44.7766 ROUGE2: 22.3407 ROUGEL: 29.1379 TOKENS_PER_SAMPLE: 275.2
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 48
Processors per node: 2
Nodes: 1
Total accelerators: 4
Software stack: TensorRT 10.14, CUDA 13.1
Operating system: Ubuntu 24.04

MLPerf Inference / llama2-70b-99.9Server · v6.1 · Tokens/s

4 × NVIDIA RTX PRO 6000 Blackwell Server Edition

14,333.30^* Tokens/s

Accuracy target / result: ROUGE1: 44.7151 ROUGE2: 22.3389 ROUGEL: 29.0874 TOKENS_PER_SAMPLE: 274.6
Weight precision: fp4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 48
Processors per node: 2
Nodes: 1
Total accelerators: 4
Software stack: TensorRT 10.14, CUDA 13.1
Operating system: Ubuntu 24.04

Processing & platform

Processor: Intel(R) Xeon(R) 6747P
Graphics: NVIDIA RTX PRO 6000 Blackwell Server Edition

Connectivity & software

Operating system: Ubuntu 24.04

System configuration

Reported host core count: 48
Processors per node: 2
Accelerator count: 4
Nodes: 1
Software stack: TensorRT 10.14, CUDA 13.1
Submission division: closed

* Provisional data