chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 11 Packages Projects Releases Wiki Activity

simplify the control logic for using shared experts fusion (#5504)

Browse Source
This commit is contained in:
Xiaoyu Zhang
2025-04-20 04:17:35 +08:00
committed by GitHub
parent bf86c5e990
commit d58e354472
16 changed files with 69 additions and 54 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
python/sglang/srt/model_executor/model_runner.py
View File

@@ -163,7 +163,6 @@ class ModelRunner:
"debug_tensor_dump_output_folder": server_args.debug_tensor_dump_output_folder,
"debug_tensor_dump_inject": server_args.debug_tensor_dump_inject,
"n_share_experts_fusion": server_args.n_share_experts_fusion,
"disable_shared_experts_fusion": server_args.disable_shared_experts_fusion,
"disable_chunked_prefix_cache": server_args.disable_chunked_prefix_cache,
"use_mla_backend": self.use_mla_backend,
}