4 × Intel(R) Arc(TM) Pro B70

Tested system

Processor: Intel(R) Xeon(R) 6731P
Graphics: Intel(R) Arc(TM) Pro B70
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 × Intel(R) Arc(TM) Pro B70

2,063.93^* Tokens/s

Accuracy target / result: exact_match: 83.780
Weight precision: MXFP4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

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

4 × Intel(R) Arc(TM) Pro B70

1,351.13^* Tokens/s

Accuracy target / result: exact_match: 84.186
Weight precision: MXFP4
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

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

4 × Intel(R) Arc(TM) Pro B70

2,484.14^* Tokens/s

Accuracy target / result: ROUGE1: 44.585 ROUGE2: 22.2008 ROUGEL: 28.8241 TOKENS_PER_SAMPLE: 290.8
Weight precision: INT8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

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

4 × Intel(R) Arc(TM) Pro B70

1,759.95^* Tokens/s

Accuracy target / result: ROUGE1: 44.591 ROUGE2: 22.2044 ROUGEL: 28.8277 TOKENS_PER_SAMPLE: 290.8
Weight precision: INT8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

MLPerf Inference / llama3_1-8bOffline · v6.1 · Tokens/s

4 × Intel(R) Arc(TM) Pro B70

4,596.62^* Tokens/s

Accuracy target / result: ROUGE1: 38.8447 ROUGE2: 16.0909 ROUGEL: 24.6442 ROUGELSUM: 35.9477 GEN_LEN: 8127563
Weight precision: INT8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

MLPerf Inference / llama3_1-8bServer · v6.1 · Tokens/s

4 × Intel(R) Arc(TM) Pro B70

4,208.73^* Tokens/s

Accuracy target / result: ROUGE1: 38.8128 ROUGE2: 16.082 ROUGEL: 24.6384 ROUGELSUM: 35.9155 GEN_LEN: 8126189
Weight precision: INT8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

MLPerf Inference / whisperOffline · v6.1 · Samples/s

4 × Intel(R) Arc(TM) Pro B70

5,854.85^* Samples/s

Accuracy target / result: ACCURACY: 97.49980776624375
Weight precision: INT8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 32
Processors per node: 1
Nodes: 1
Total accelerators: 4
Software stack: PyTorch
Operating system: Ubuntu 24.04.4 LTS

Processing & platform

Processor: Intel(R) Xeon(R) 6731P
Graphics: Intel(R) Arc(TM) Pro B70

Connectivity & software

Operating system: Ubuntu 24.04.4 LTS

System configuration

Reported host core count: 32
Processors per node: 1
Accelerator count: 4
Nodes: 1
Software stack: PyTorch
Submission division: closed

* Provisional data