mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
13 lines
352 B
YAML
13 lines
352 B
YAML
max_model_len: 32768
|
|
max_num_seqs: 16
|
|
gpu_memory_utilization: 0.9
|
|
kv_cache_ratio: 0.9
|
|
tensor_parallel_size: 1
|
|
data_parallel_size: 8
|
|
splitwise_role: prefill
|
|
cache_queue_port: 55664
|
|
engine_worker_queue_port: 6677
|
|
num_gpu_blocks_override: 1024
|
|
cache_transfer_protocol: "rdma"
|
|
rdma_comm_ports: "7671,7672,7673,7674,7675,7676,7677,7678"
|
|
pd_comm_port: "2334" |