LeaseWeb amd-epyc-9334-genoa-dual Llama-2-7b Q4_K_M TTFT s
4800.0 ms
Fields
id
benchmarks/lw-epyc9334-dual-llama27b-q4-ttftMetric
ttft_ms
Value
4,800
Unit
ms
Model
Llama-2-7b
Methodology
LeaseWeb benchmark (HPE ProLiant DL385 Gen11, dual EPYC 9334 32C Zen4, 64GB DDR5; L4 reference single GPU). llama-bench + OpenLLM/llmperf. 512-token prompt, 128 generated, 24 CPU threads for LLM runs. TTFT s.
Measured by
third_party
Reproducible
yes
Thermal state
unknown
notes
Dual-socket EPYC 9334 (2x32C) unless L4. LLM at 24 threads.
Notes
Dual-socket EPYC 9334 (2x32C) unless L4. LLM at 24 threads.