apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

check max_logprobs (#5018)

Browse Source
This commit is contained in:
chen
2025-11-14 17:18:06 +08:00
committed by GitHub
parent 249feca65a
commit 544ea9cbc2
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/engine/args_utils.py
View File

@@ -489,7 +489,7 @@ class EngineArgs:
raise NotImplementedError("processed_logprobs not support in speculative.")
if self.speculative_config is not None and self.max_logprobs == -1:
raise NotImplementedError("max_logprobs=-1 not support in speculative.")
if not envs.FD_USE_GET_SAVE_OUTPUT_V1:
if not envs.FD_USE_GET_SAVE_OUTPUT_V1 and (self.max_logprobs == -1 or self.max_logprobs > 20):
self.max_logprobs = 20
console_logger.warning("Set max_logprobs=20 when FD_USE_GET_SAVE_OUTPUT_V1=0")
if self.max_logprobs == -1 and not envs.ENABLE_V1_KVCACHE_SCHEDULER: