GeekRadar / The comparison desk
COMPARE.
Tested systems16 × 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
137,067* Tokens/s
- Accuracy target / result
- exact_match: 81.33546034639927 TOKENS_PER_SAMPLE: 3825.4619416590704
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
MLPerf Inference / deepseek-r1Server · v6.1 · Tokens/s
16 × NVIDIA B300-SXM-270GB
88,915.50* Tokens/s
- Accuracy target / result
- exact_match: 81.28988149498633 TOKENS_PER_SAMPLE: 3794.9457611668186
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
MLPerf Inference / gpt-oss-120bOffline · v6.1 · Tokens/s
16 × NVIDIA B300-SXM-270GB
220,527* Tokens/s
- Accuracy target / result
- exact_match: 83.296
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
MLPerf Inference / gpt-oss-120bServer · v6.1 · Tokens/s
16 × NVIDIA B300-SXM-270GB
214,774* Tokens/s
- Accuracy target / result
- exact_match: 83.572
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
MLPerf Inference / llama2-70b-99.9Offline · v6.1 · Tokens/s
16 × NVIDIA B300-SXM-270GB
227,570* Tokens/s
- Accuracy target / result
- ROUGE1: 44.8082 ROUGE2: 22.3989 ROUGEL: 29.1755 TOKENS_PER_SAMPLE: 272.1
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
MLPerf Inference / llama2-70b-99.9Server · v6.1 · Tokens/s
16 × NVIDIA B300-SXM-270GB
179,597* Tokens/s
- Accuracy target / result
- ROUGE1: 44.7969 ROUGE2: 22.3691 ROUGEL: 29.184 TOKENS_PER_SAMPLE: 272.7
- Weight precision
- fp4
- Division
- closed
- System type
- datacenter
- Availability
- available
- Cores per processor
- 128
- Processors per node
- 2
- Nodes
- 2
- Total accelerators
- 16
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
Processing & platform
- Processor
- Intel(R) Xeon(R) 6776P
- Graphics
- NVIDIA B300-SXM-270GB
Connectivity & software
- Operating system
- Ubuntu 24.04.4 LTS, Linux 6.17.0-35-generic
System configuration
- Reported host core count
- 128
- Processors per node
- 2
- Accelerator count
- 16
- Nodes
- 2
- Software stack
- TensorRT 10.14.1.48; TensorRT-LLM 1.3.0rc0
- Submission division
- closed
0 selected Compare 0
* Provisional data