mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-12-24 13:28:13 +08:00
10 lines
219 B
YAML
10 lines
219 B
YAML
enable_chunked_prefill: True
|
|
max_model_len: 131072
|
|
max_num_seqs: 16
|
|
kv_cache_ratio: 0.75
|
|
tensor_parallel_size: 8
|
|
max_num_batched_tokens: 4096
|
|
max_num_partial_prefills: 3
|
|
max_long_partial_prefills: 3
|
|
quantization: wint4
|