MLPerf v6.1 Dell R770 2x Xeon 6787P: llama3.1-8b INT8 Offline aggregate tok/s

MLPerf Inference v6.1 closed division, CPU-only (summary.csv accelerator_model_name is N/A). Values copied from summary.csv, not from a per-system mlperf_log_summary.txt. AGGREGATE throughput for the whole node, NOT per-request decode and NOT per-socket. Prompt length, output length, and batch size are not in the summary row. weight_data_types INT8. Framework column is PyTorch; the Dell notes field additionally names vLLM with an FP8 KV cache. Scenario: Offline. System: Dell PowerEdge R770, 2x Xeon 6787P, 16x64GB DDR5-6400. Best-documented CPU row in the file: summary.csv notes say 'Dell PowerEdge R770. vLLM, FP8 KV Cache'. 16 DIMMs on an 8-channel part is 1 DPC, fully populated. 8 channels, not 12 — do not read this as a 6980P number. INT8 weights, FP8 KV. measured_by vendor: these are submitter results inside the MLCommons closed division (Intel, Dell, Supermicro, Quanta, Red Hat, Cisco), not an independent re-run.

Best-documented CPU row in the file: summary.csv notes say 'Dell PowerEdge R770. vLLM, FP8 KV Cache'. 16 DIMMs on an 8-channel part is 1 DPC, fully populated. 8 channels, not 12 — do not read this as a 6980P number. INT8 weights, FP8 KV.