minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

call check_quantized_moe_compatibility after initialize (#13876)

Browse Source
This commit is contained in:
Chunyuan WU
2025-12-13 13:32:19 +08:00
committed by GitHub
parent bf17e769fe
commit 2a39cfe0ff
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/model_executor/model_runner.py
+1 -1
View File
@@ -323,7 +323,6 @@ class ModelRunner:
# Model-specific adjustment # Model-specific adjustment
self.model_specific_adjustment() self.model_specific_adjustment()
self.check_quantized_moe_compatibility()
# Set the global server_args in the scheduler process # Set the global server_args in the scheduler process
set_global_server_args_for_scheduler(server_args) set_global_server_args_for_scheduler(server_args)
@@ -355,6 +354,7 @@ class ModelRunner:
# Initialize the model runner # Initialize the model runner
self.initialize(min_per_gpu_memory) self.initialize(min_per_gpu_memory)
self.check_quantized_moe_compatibility()
# Temporary cached values # Temporary cached values
self.support_pp = ( self.support_pp = (