8 × AMD Instinct MI350X 288GB HBM3e

Tested system

Processor: AMD EPYC 9575F
Graphics: AMD Instinct MI350X 288GB HBM3e
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 / dlrm-v3Offline · v6.1 · Samples/s

8 × AMD Instinct MI350X 288GB HBM3e

10,891.10^* Samples/s

Accuracy target / result: DLRM_NE: 0.8677294228203597 DLRM_ACC: 0.6964170290002015 DLRM_AUC: 0.7862871466027813
Weight precision: fp16,bf16,fp8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 1
Total accelerators: 8
Software stack: PyTorch 2.10.0+rocm7.2.3, Triton 3.6.0, ROCm 7.2.3
Operating system: Ubuntu 24.04 LTS (Noble Numbat), kernel 6.8.0-38-generic

MLPerf Inference / dlrm-v3Server · v6.1 · Queries/s

8 × AMD Instinct MI350X 288GB HBM3e

9,297.78^* Queries/s

Accuracy target / result: DLRM_NE: 0.867728327062296 DLRM_ACC: 0.696416918732384 DLRM_AUC: 0.786287540637186
Weight precision: fp16,bf16,fp8
Division: closed
System type: datacenter
Availability: available
Cores per processor: 64
Processors per node: 2
Nodes: 1
Total accelerators: 8
Software stack: PyTorch 2.10.0+rocm7.2.3, Triton 3.6.0, ROCm 7.2.3
Operating system: Ubuntu 24.04 LTS (Noble Numbat), kernel 6.8.0-38-generic

Processing & platform

Processor: AMD EPYC 9575F
Graphics: AMD Instinct MI350X 288GB HBM3e

Connectivity & software

Operating system: Ubuntu 24.04 LTS (Noble Numbat), kernel 6.8.0-38-generic

System configuration

Reported host core count: 64
Processors per node: 2
Accelerator count: 8
Nodes: 1
Software stack: PyTorch 2.10.0+rocm7.2.3, Triton 3.6.0, ROCm 7.2.3
Submission division: closed

* Provisional data