data/mllm_text_generation/chat/A100-SXM4-40GB/llava-hf/llama3-llava-next-8b-hf/bs32+tp2+pp1.json

{
  "Model": "llava-hf/llama3-llava-next-8b-hf",
  "GPU": "NVIDIA A100-SXM4-40GB",
  "TP": 2,
  "PP": 1,
  "Energy/req (J)": 193.8074747286492,
  "Avg TPOT (s)": 0.06319119201205099,
  "Token tput (tok/s)": 468.0953448723822,
  "Avg Output Tokens": 225.382,
  "Avg BS (reqs)": 31.76177072671443,
  "Max BS (reqs)": 32
}