EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Clean up wrapper in flashinfer backend (#2638)

Browse Source
This commit is contained in:
Lianmin Zheng
2024-12-29 00:45:57 -08:00
committed by GitHub
parent fd34f2da35
commit 3815b23ccb
12 changed files with 197 additions and 94 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/sglang/srt/server.py
View File

@@ -503,7 +503,7 @@ def launch_engine(
)
scheduler_infos.append(data)
# Assume all schedulers have same max_total_num_tokens
# Assume all schedulers have same scheduler_info
scheduler_info = scheduler_infos[0]
@@ -890,7 +890,7 @@ class Runtime:
using the commond line interface.
It is mainly used for the frontend language.
You should use the Engine class if you want to do normal offline processing.
You should use the Engine class above if you want to do normal offline processing.
"""
def __init__(