diff --git a/python/sglang/srt/model_executor/model_runner.py b/python/sglang/srt/model_executor/model_runner.py
index a0272029aa2d..324d5f314790 100644
--- a/python/sglang/srt/model_executor/model_runner.py
+++ b/python/sglang/srt/model_executor/model_runner.py
@@ -538,6 +538,7 @@ def initialize(self, min_per_gpu_memory: float):
             draft_model_config = ModelConfig.from_server_args(
                 server_args,
                 model_path=(server_args.speculative_draft_model_path),
+                model_revision=server_args.speculative_draft_model_revision,
                 is_draft_model=True,
             )
 
diff --git a/python/sglang/srt/server_args.py b/python/sglang/srt/server_args.py
index 792201428cc3..5720ae3fa0be 100644
--- a/python/sglang/srt/server_args.py
+++ b/python/sglang/srt/server_args.py
@@ -1610,6 +1610,12 @@ def _handle_hicache(self):
             )
 
     def _handle_speculative_decoding(self):
+        if (
+            self.speculative_draft_model_path is not None
+            and self.speculative_draft_model_revision is None
+        ):
+            self.speculative_draft_model_revision = "main"
+
         if self.speculative_algorithm == "NEXTN":
             self.speculative_algorithm = "EAGLE"
 
@@ -1658,6 +1664,7 @@ def _handle_speculative_decoding(self):
             ]:
                 if self.speculative_draft_model_path is None:
                     self.speculative_draft_model_path = self.model_path
+                    self.speculative_draft_model_revision = self.revision
                 else:
                     logger.warning(
                         "DeepSeek MTP does not require setting speculative_draft_model_path."