GeekRadar / The comparison desk
COMPARE.
Tested systems
HPE

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

Tested system

Processor
Intel(R) Xeon(R) 6787P
Graphics
NVIDIA RTX PRO 6000 Blackwell Server Edition
Accelerator count
8
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

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

15,256* Tokens/s

Accuracy target / result
exact_match: 82.714
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / gpt-oss-120bServer · v6.1 · Tokens/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

14,581.70* Tokens/s

Accuracy target / result
exact_match: 82.341
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / llama2-70b-99.9Offline · v6.1 · Tokens/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

29,284.80* Tokens/s

Accuracy target / result
ROUGE1: 44.7216 ROUGE2: 22.3081 ROUGEL: 29.0908 TOKENS_PER_SAMPLE: 274.8
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / llama2-70b-99.9Server · v6.1 · Tokens/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

28,643.30* Tokens/s

Accuracy target / result
ROUGE1: 44.7706 ROUGE2: 22.352 ROUGEL: 29.1377 TOKENS_PER_SAMPLE: 272.0
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / llama3.1-8bOffline · v6.1 · Tokens/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

50,965.30* Tokens/s

Accuracy target / result
ROUGE1: 38.7974 ROUGE2: 16.0868 ROUGEL: 24.5879 ROUGELSUM: 35.7199 GEN_LEN: 8172782
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / llama3.1-8bServer · v6.1 · Tokens/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

49,523.30* Tokens/s

Accuracy target / result
ROUGE1: 38.7821 ROUGE2: 16.0772 ROUGEL: 24.5857 ROUGELSUM: 35.7074 GEN_LEN: 8172428
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS
MLPerf Inference / whisperOffline · v6.1 · Samples/s

8 × NVIDIA RTX PRO 6000 Blackwell Server Edition

19,278.30* Samples/s

Accuracy target / result
ACCURACY: 97.88345689351618
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
172
Processors per node
2
Nodes
1
Total accelerators
8
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Operating system
Ubuntu 24.04.4 LTS

Processing & platform

Processor
Intel(R) Xeon(R) 6787P
Graphics
NVIDIA RTX PRO 6000 Blackwell Server Edition

Connectivity & software

Operating system
Ubuntu 24.04.4 LTS

System configuration

Reported host core count
172
Processors per node
2
Accelerator count
8
Nodes
1
Software stack
TensorRT 10.16.1.11, CUDA 13.2
Submission division
closed

* Provisional data