{
  "Model": "codellama/CodeLlama-70b-hf",
  "GPU": "NVIDIA H100 80GB HBM3",
  "TP": 4,
  "PP": 1,
  "Energy/req (J)": 110.17968289892558,
  "Avg TPOT (s)": 0.16014937761515807,
  "Token tput (tok/s)": 941.5781689523762,
  "Avg Output Tokens": 82.2140243902439,
  "Avg BS (reqs)": 252.8793324775353,
  "Max BS (reqs)": 256
}