Files
FastDeploy/benchmarks/yaml/x1-32k-wint8-p800-tp4.yaml
2025-06-29 23:29:37 +00:00

7 lines
136 B
YAML

max_model_len: 32768
max_num_seqs: 8
gpu_memory_utilization: 0.9
tensor_parallel_size: 4
quantization: wint8
reasoning_parser: ernie-x1