diff --git a/python/sglang/srt/model_executor/cuda_graph_runner.py b/python/sglang/srt/model_executor/cuda_graph_runner.py
index 033111e45..75f085ae4 100644
--- a/python/sglang/srt/model_executor/cuda_graph_runner.py
+++ b/python/sglang/srt/model_executor/cuda_graph_runner.py
@@ -269,10 +269,10 @@ class CudaGraphRunner:
             raise Exception(
                 f"Capture cuda graph failed: {e}\n"
                 "Possible solutions:\n"
-                "1. disable cuda graph by --disable-cuda-graph\n"
-                "2. set --mem-fraction-static to a smaller value (e.g., 0.8 or 0.7)\n"
+                "1. set --mem-fraction-static to a smaller value (e.g., 0.8 or 0.7)\n"
+                "2. set --cuda-graph-max-bs to a smaller value (e.g., 32)\n"
                 "3. disable torch compile by not using --enable-torch-compile\n"
-                "4. set --cuda-graph-max-bs to a smaller value (e.g., 32)\n"
+                "4. disable cuda graph by --disable-cuda-graph\n"
                 "Open an issue on GitHub https://github.com/sgl-project/sglang/issues/new/choose \n"
             )
 
diff --git a/python/sglang/srt/model_executor/model_runner.py b/python/sglang/srt/model_executor/model_runner.py
index e7fae3a68..7c9ec61da 100644
--- a/python/sglang/srt/model_executor/model_runner.py
+++ b/python/sglang/srt/model_executor/model_runner.py
@@ -924,6 +924,12 @@ class ModelRunner:
             return
 
         if self.server_args.disable_cuda_graph:
+            logger.warning(
+                "\n\nCUDA Graph is DISABLED.\n"
+                "This will cause significant performance degradation.\n"
+                "CUDA Graph should almost never be disabled in most usage scenarios.\n"
+                "If you encounter OOM issues, please try setting --mem-fraction-static to a lower value (such as 0.8 or 0.7) instead of disabling CUDA Graph.\n"
+            )
             return
 
         tic = time.time()
diff --git a/python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py b/python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py
index d3ad50060..19866291b 100644
--- a/python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py
+++ b/python/sglang/srt/speculative/eagle_draft_cuda_graph_runner.py
@@ -84,10 +84,10 @@ class EAGLEDraftCudaGraphRunner:
             raise Exception(
                 f"Capture cuda graph failed: {e}\n"
                 "Possible solutions:\n"
-                "1. disable cuda graph by --disable-cuda-graph\n"
-                "2. set --mem-fraction-static to a smaller value (e.g., 0.8 or 0.7)\n"
-                "3. disable torch compile by not using --enable-torch-compile\n"
-                "4. specify --dtype to the same dtype (e.g. bfloat16)\n"
+                "1. set --mem-fraction-static to a smaller value (e.g., 0.8 or 0.7)\n"
+                "2. disable torch compile by not using --enable-torch-compile\n"
+                "3. specify --dtype to the same dtype (e.g. bfloat16)\n"
+                "4. disable cuda graph by --disable-cuda-graph\n"
                 "Open an issue on GitHub https://github.com/sgl-project/sglang/issues/new/choose \n"
             )