Jae-Won Chung
New leaderboard prototype
b10121d
raw
history blame contribute delete
315 Bytes
{
"Model": "codellama/CodeLlama-7b-hf",
"GPU": "NVIDIA A100-SXM4-40GB",
"TP": 2,
"PP": 1,
"Energy/req (J)": 21.225724206645324,
"Avg TPOT (s)": 0.14097590539196214,
"Token tput (tok/s)": 1677.1309064039083,
"Avg Output Tokens": 99.925,
"Avg BS (reqs)": 313.62927414517094,
"Max BS (reqs)": 320
}