GeekRadar / The comparison desk
COMPARE.
Tested systems
Nebius

72 × NVIDIA GB300

Tested system

Processor
NVIDIA Grace CPU
Graphics
NVIDIA GB300
Accelerator count
72
Nodes
18

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

72 × NVIDIA GB300

689,961* Tokens/s

Accuracy target / result
exact_match: 80.94804010938924 TOKENS_PER_SAMPLE: 3831.7297174111213
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
144
Processors per node
2
Nodes
18
Total accelerators
72
Software stack
TensorRT 10.14, CUDA 13.1, cuDNN 9.17, TensorRT-LLM feat/1.2-mlpinf, NVIDIA Dynamo mlperf-v6.0-dynamo-v0.8.0, vLLM CentML:mlperf-inf-mm-q3vl-v6.0
Operating system
Ubuntu 24.04
MLPerf Inference / deepseek-r1Server · v6.1 · Tokens/s

72 × NVIDIA GB300

603,023* Tokens/s

Accuracy target / result
exact_match: 80.67456700091158 TOKENS_PER_SAMPLE: 3850.7272105742936
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
144
Processors per node
2
Nodes
18
Total accelerators
72
Software stack
TensorRT 10.14, CUDA 13.1, cuDNN 9.17, TensorRT-LLM feat/1.2-mlpinf, NVIDIA Dynamo mlperf-v6.0-dynamo-v0.8.0, vLLM CentML:mlperf-inf-mm-q3vl-v6.0
Operating system
Ubuntu 24.04
MLPerf Inference / gpt-oss-120bOffline · v6.1 · Tokens/s

72 × NVIDIA GB300

1,186,750* Tokens/s

Accuracy target / result
exact_match: 82.905
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
144
Processors per node
2
Nodes
18
Total accelerators
72
Software stack
TensorRT 10.14, CUDA 13.1, cuDNN 9.17, TensorRT-LLM feat/1.2-mlpinf, NVIDIA Dynamo mlperf-v6.0-dynamo-v0.8.0, vLLM CentML:mlperf-inf-mm-q3vl-v6.0
Operating system
Ubuntu 24.04
MLPerf Inference / gpt-oss-120bServer · v6.1 · Tokens/s

72 × NVIDIA GB300

1,122,490* Tokens/s

Accuracy target / result
exact_match: 83.174
Weight precision
fp4
Division
closed
System type
datacenter
Availability
available
Cores per processor
144
Processors per node
2
Nodes
18
Total accelerators
72
Software stack
TensorRT 10.14, CUDA 13.1, cuDNN 9.17, TensorRT-LLM feat/1.2-mlpinf, NVIDIA Dynamo mlperf-v6.0-dynamo-v0.8.0, vLLM CentML:mlperf-inf-mm-q3vl-v6.0
Operating system
Ubuntu 24.04

Processing & platform

Processor
NVIDIA Grace CPU
Graphics
NVIDIA GB300

Connectivity & software

Operating system
Ubuntu 24.04

System configuration

Reported host core count
144
Processors per node
2
Accelerator count
72
Nodes
18
Software stack
TensorRT 10.14, CUDA 13.1, cuDNN 9.17, TensorRT-LLM feat/1.2-mlpinf, NVIDIA Dynamo mlperf-v6.0-dynamo-v0.8.0, vLLM CentML:mlperf-inf-mm-q3vl-v6.0
Submission division
closed

* Provisional data