Files
FastDeploy/benchmarks/yaml/x1-a3b-128k-wint8-h800-tp1.yaml
2025-09-16 19:44:46 +08:00

7 lines
144 B
YAML

tensor_parallel_size: 1
max_model_len: 131072
max_num_seqs: 32
reasoning_parser: ernie_x1
tool_call_parser: ernie_x1
load_choices: "default_v1"