data/mllm_text_generation/chat/A100-SXM4-40GB/llava-hf/llama3-llava-next-8b-hf/bs32+tp1+pp1.json

{
  "Model": "llava-hf/llama3-llava-next-8b-hf",
  "GPU": "NVIDIA A100-SXM4-40GB",
  "TP": 1,
  "PP": 1,
  "Energy/req (J)": 144.41504192259845,
  "Avg TPOT (s)": 0.07198214306309311,
  "Token tput (tok/s)": 405.8532041975983,
  "Avg Output Tokens": 224.128,
  "Avg BS (reqs)": 31.773305634900026,
  "Max BS (reqs)": 32
}