diff --git a/megatron/rl/rl_utils.py b/megatron/rl/rl_utils.py
index 8a564315dc3..4d018217cec 100644
--- a/megatron/rl/rl_utils.py
+++ b/megatron/rl/rl_utils.py
@@ -1973,8 +1973,8 @@ def megatron_rl_inference_mode(
 
     logger.debug(f"[{dist.get_rank()}] Entering inference mode")
 
-    # Change cudagraph scope for inference (empty list = full-layer capture)
-    model[0].config.cuda_graph_scope = []
+    # Set cudagraph scope for inference.
+    model[0].config.cuda_graph_scope = args.cuda_graph_scope
     model[0].config.cuda_graph_impl = "local"
 
     # If we get a lower precision wrapper, we go one object deeper.
@@ -2031,7 +2031,8 @@ def megatron_rl_inference_mode(
         # Reset drop_and_pad leaked from inference decode
         set_decode_expert_padding(unwrap_model(model[0]), set_to=False)
 
-        # Restore partial capture cudagraph scope for training if this is MoE
+        # Restore cudagraph scope for training.
+        # MoE partial capture requires specific scopes that aren't user-facing.
         if args.num_experts is not None:
             model[0].config.cuda_graph_scope = [
                 CudaGraphScope.mamba,
@@ -2039,6 +2040,10 @@ def megatron_rl_inference_mode(
                 CudaGraphScope.moe_router,
                 CudaGraphScope.moe_preprocess,
             ]
+        else:
+            model[0].config.cuda_graph_scope = [
+                s for s in args.cuda_graph_scope if s != CudaGraphScope.full_iteration_inference
+            ]
 
         # Switch MoE layers to partial CUDA graph capture for training
         if args.rl_training_cuda_graphs and args.num_experts is not None: