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

[Feature] Support DeepEP Low Latency (#4767)

Browse Source 
Co-authored-by: sleepcoo <sleepcoo@gmail.com>
Co-authored-by: laixinn <xielx@shanghaitech.edu.cn>
Co-authored-by: ch-wan <cwan39@gatech.edu>
This commit is contained in:
Jinyan Chen
2025-04-02 00:23:25 +08:00
committed by GitHub
parent 87fafa0105
commit 23c764b18a
8 changed files with 448 additions and 238 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
python/sglang/srt/model_executor/model_runner.py
View File

@@ -147,6 +147,7 @@ class ModelRunner:
"enable_dp_attention": server_args.enable_dp_attention,
"enable_ep_moe": server_args.enable_ep_moe,
"enable_deepep_moe": server_args.enable_deepep_moe,
"deepep_mode": server_args.deepep_mode,
"device": server_args.device,
"speculative_accept_threshold_single": server_args.speculative_accept_threshold_single,
"speculative_accept_threshold_acc": server_args.speculative_accept_threshold_acc,
@@ -272,7 +273,7 @@ class ModelRunner:
server_args.disable_radix_cache = True
if server_args.enable_deepep_moe:
logger.info("DeepEP is turned on.")
logger.info(f"DeepEP is turned on. DeepEP mode: {server_args.deepep_mode}")
def init_torch_distributed(self):
logger.info("Init torch distributed begin.")