Skip to content

GitLab

Unverified Commit 97cd38e5 authored by Cheng Wan's avatar Cheng Wan Committed by GitHub
Browse files

Skip weight loading in deepgemm compilation (#11312)

parent 3c06b673
Show whitespace changes
Inline Side-by-side
Showing with 3 additions and 0 deletions
python/sglang/compile_deep_gemm.py
View file @ 97cd38e5
...@@ -141,6 +141,9 @@ def refine_server_args(server_args: ServerArgs, compile_args: CompileArgs): ...@@ -141,6 +141,9 @@ def refine_server_args(server_args: ServerArgs, compile_args: CompileArgs):
server_args.enable_torch_compile = False server_args.enable_torch_compile = False
print(f"Disable CUDA Graph and Torch Compile to save time...") print(f"Disable CUDA Graph and Torch Compile to save time...")
server_args.load_format = "dummy"
print(f"Set load format to dummy to save time...")
# Set watchdog timeout to compile_args.timeout because compilation will take a long time # Set watchdog timeout to compile_args.timeout because compilation will take a long time
server_args.watchdog_timeout = compile_args.timeout server_args.watchdog_timeout = compile_args.timeout
server_args.warmups = "compile-deep-gemm" server_args.warmups = "compile-deep-gemm"
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment