defalut FD_ENABLE_TOP_P_ONE_OPT=0

ckl117 · ckl117 · commit f41e457fdee6 · 2026-05-26T22:23:11.000+08:00
diff --git a/fastdeploy/envs.py b/fastdeploy/envs.py
@@ -290,7 +290,7 @@ def _validate_split_kv_size(value: int) -> int:
     # Whether to enable FP8 quantization with pow2scale.
     "FD_FP8_QUANT_WITH_POW2SCALE": lambda: bool(int(os.getenv("FD_FP8_QUANT_WITH_POW2SCALE", "0"))),
     # Whether to enable top_p=1.0 optimization.
-    "FD_ENABLE_TOP_P_ONE_OPT": lambda: bool(int(os.getenv("FD_ENABLE_TOP_P_ONE_OPT", "1"))),
+    "FD_ENABLE_TOP_P_ONE_OPT": lambda: bool(int(os.getenv("FD_ENABLE_TOP_P_ONE_OPT", "0"))),
 }
 
 

Original file line number	Diff line number	Diff line change
`@@ -290,7 +290,7 @@ def _validate_split_kv_size(value: int) -> int:`
`290`	`290`	`# Whether to enable FP8 quantization with pow2scale.`
`291`	`291`	`"FD_FP8_QUANT_WITH_POW2SCALE": lambda: bool(int(os.getenv("FD_FP8_QUANT_WITH_POW2SCALE", "0"))),`
`292`	`292`	`# Whether to enable top_p=1.0 optimization.`
`293`		`- "FD_ENABLE_TOP_P_ONE_OPT": lambda: bool(int(os.getenv("FD_ENABLE_TOP_P_ONE_OPT", "1"))),`
	`293`	`+ "FD_ENABLE_TOP_P_ONE_OPT": lambda: bool(int(os.getenv("FD_ENABLE_TOP_P_ONE_OPT", "0"))),`
`294`	`294`	`}`
`295`	`295`
`296`	`296`