data/mllm_text_generation/chat/A100-SXM4-40GB/llava-hf/llama3-llava-next-8b-hf/bs128+tp2+pp1.json

{
  "Model": "llava-hf/llama3-llava-next-8b-hf",
  "GPU": "NVIDIA A100-SXM4-40GB",
  "TP": 2,
  "PP": 1,
  "Energy/req (J)": 167.65754307648515,
  "Avg TPOT (s)": 0.13820982397168496,
  "Token tput (tok/s)": 600.3811225799612,
  "Avg Output Tokens": 227.53,
  "Avg BS (reqs)": 121.77866880513231,
  "Max BS (reqs)": 128
}