data/mllm_text_generation/chat/A100-SXM4-40GB/llava-hf/llama3-llava-next-8b-hf/bs64+tp1+pp1.json

{
  "Model": "llava-hf/llama3-llava-next-8b-hf",
  "GPU": "NVIDIA A100-SXM4-40GB",
  "TP": 1,
  "PP": 1,
  "Energy/req (J)": 130.4830497702924,
  "Avg TPOT (s)": 0.1043024722774119,
  "Token tput (tok/s)": 490.58067954477787,
  "Avg Output Tokens": 228.414,
  "Avg BS (reqs)": 62.94636194029851,
  "Max BS (reqs)": 64
}