31 lines
827 B
JSON
31 lines
827 B
JSON
[
|
|
{
|
|
"test_name": "latency_llama8B_tp1",
|
|
"environment_variables": {
|
|
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
|
|
"VLLM_CPU_KVCACHE_SPACE": 40
|
|
},
|
|
"parameters": {
|
|
"model": "meta-llama/Meta-Llama-3.1-8B-Instruct",
|
|
"tensor_parallel_size": 1,
|
|
"load_format": "dummy",
|
|
"num_iters_warmup": 5,
|
|
"num_iters": 15
|
|
}
|
|
},
|
|
{
|
|
"test_name": "latency_llama8B_tp4",
|
|
"environment_variables": {
|
|
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
|
|
"VLLM_CPU_KVCACHE_SPACE": 40
|
|
},
|
|
"parameters": {
|
|
"model": "meta-llama/Meta-Llama-3.1-8B-Instruct",
|
|
"tensor_parallel_size": 4,
|
|
"load_format": "dummy",
|
|
"num_iters_warmup": 5,
|
|
"num_iters": 15
|
|
}
|
|
}
|
|
]
|