From b6b6268ccf1d992cac417b995fd250281c17912a Mon Sep 17 00:00:00 2001 From: Yineng Zhang Date: Tue, 8 Jul 2025 22:03:09 -0700 Subject: [PATCH] Revert "Embedding parallel by attn_tp (#7623)" (#7880) --- python/sglang/srt/models/deepseek_v2.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/python/sglang/srt/models/deepseek_v2.py b/python/sglang/srt/models/deepseek_v2.py index 1784ee132..18c408bc7 100644 --- a/python/sglang/srt/models/deepseek_v2.py +++ b/python/sglang/srt/models/deepseek_v2.py @@ -1938,7 +1938,7 @@ class DeepseekV2Model(nn.Module): self.embed_tokens = VocabParallelEmbedding( config.vocab_size, config.hidden_size, - use_attn_tp_group=True, + enable_tp=not global_server_args_dict["enable_dp_attention"], ) self.alt_stream = torch.cuda.Stream() if _is_cuda else None self.layers = nn.ModuleList(