diff --git a/python/sglang/srt/layers/moe/token_dispatcher/deepep.py b/python/sglang/srt/layers/moe/token_dispatcher/deepep.py index 7c8704439..17ca310ab 100644 --- a/python/sglang/srt/layers/moe/token_dispatcher/deepep.py +++ b/python/sglang/srt/layers/moe/token_dispatcher/deepep.py @@ -307,6 +307,7 @@ class _DeepEPDispatcherImplBase: self.deepep_mode = deepep_mode self.params_bytes = 2 + # A large value will lead to large memory occupation, thus users should change it accordingly self.num_max_dispatch_tokens_per_rank = get_int_env_var( "SGLANG_DEEPEP_NUM_MAX_DISPATCH_TOKENS_PER_RANK", 128 )