data/mllm_text_generation/chat/A100-SXM4-40GB/llava-hf/llama3-llava-next-8b-hf/bs64+tp2+pp1.json

{
  "Model": "llava-hf/llama3-llava-next-8b-hf",
  "GPU": "NVIDIA A100-SXM4-40GB",
  "TP": 2,
  "PP": 1,
  "Energy/req (J)": 171.24595824069502,
  "Avg TPOT (s)": 0.09632233914192728,
  "Token tput (tok/s)": 552.6775236368885,
  "Avg Output Tokens": 224.208,
  "Avg BS (reqs)": 62.94396961063628,
  "Max BS (reqs)": 64
}