mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
10 lines
202 B
YAML
10 lines
202 B
YAML
max_model_len: 32768
|
|
max_num_seqs: 96
|
|
gpu_memory_utilization: 0.85
|
|
kv_cache_ratio: 0.71
|
|
tensor_parallel_size: 4
|
|
quantization: wint4
|
|
graph_optimization_config:
|
|
use_cudagraph: True
|
|
graph_opt_level: 2
|