From e8a2327d523ce646edf400a2c6da647ca7d8c645 Mon Sep 17 00:00:00 2001
From: Lianmin Zheng <lianminzheng@gmail.com>
Date: Fri, 7 Jun 2024 19:49:18 -0700
Subject: [PATCH] Update version to 0.1.17 (#515)

---
 python/pyproject.toml                                | 2 +-
 python/sglang/__init__.py                            | 2 +-
 python/sglang/global_config.py                       | 4 ++--
 python/sglang/srt/managers/controller/infer_batch.py | 1 +
 4 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/python/pyproject.toml b/python/pyproject.toml
index 343b555f3..a932a263e 100644
--- a/python/pyproject.toml
+++ b/python/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "sglang"
-version = "0.1.16"
+version = "0.1.17"
 description = "A structured generation langauge for LLMs."
 readme = "README.md"
 requires-python = ">=3.8"
diff --git a/python/sglang/__init__.py b/python/sglang/__init__.py
index fd2bbe14d..d6303c86a 100644
--- a/python/sglang/__init__.py
+++ b/python/sglang/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "0.1.16"
+__version__ = "0.1.17"
 
 # SGL API Components
 from sglang.api import (
diff --git a/python/sglang/global_config.py b/python/sglang/global_config.py
index ba1a17b38..edfd768e1 100644
--- a/python/sglang/global_config.py
+++ b/python/sglang/global_config.py
@@ -26,8 +26,8 @@ class GlobalConfig:
         self.concate_and_append_mode = "no_adjust"
 
         # Request dependency time due to network delay
-        self.request_dependency_delay = 0.03
-        self.wait_for_new_request_delay = 0.0006
+        self.request_dependency_delay = 0.02
+        self.wait_for_new_request_delay = 0.0004
 
         # New generation token ratio estimation
         self.base_new_token_ratio = 0.4
diff --git a/python/sglang/srt/managers/controller/infer_batch.py b/python/sglang/srt/managers/controller/infer_batch.py
index 410f8a230..243a4397c 100644
--- a/python/sglang/srt/managers/controller/infer_batch.py
+++ b/python/sglang/srt/managers/controller/infer_batch.py
@@ -15,6 +15,7 @@ class ForwardMode(IntEnum):
     EXTEND = auto()
     DECODE = auto()
 
+
 class BaseFinishReason:
     def __init__(self, is_error: bool = False):
         self.is_error = is_error