diff --git a/litellm/types/utils.py b/litellm/types/utils.py
index 50e4687b5a8..0f5ccaf17ac 100644
--- a/litellm/types/utils.py
+++ b/litellm/types/utils.py
@@ -1224,6 +1224,13 @@ class Delta(SafeAttributeModel, OpenAIObject):
         annotations: Optional[List[ChatCompletionAnnotation]] = None,
         **params,
     ):
+        # Map 'reasoning' to 'reasoning_content' for providers that return
+        # delta.reasoning (e.g., Cerebras, Groq gpt-oss models).
+        # Must be done before super().__init__ to prevent 'reasoning' from
+        # leaking as an extra attribute on the parent model.
+        if reasoning_content is None and "reasoning" in params:
+            reasoning_content = params.pop("reasoning", None)
+
         super(Delta, self).__init__(**params)
         add_provider_specific_fields(self, params.get("provider_specific_fields", {}))
         self.content = content
diff --git a/tests/test_litellm/types/test_types_utils.py b/tests/test_litellm/types/test_types_utils.py
index 70344950d3e..93c234a4667 100644
--- a/tests/test_litellm/types/test_types_utils.py
+++ b/tests/test_litellm/types/test_types_utils.py
@@ -222,3 +222,30 @@ def test_chat_completion_token_logprob_invalid_top_logprobs_rejected():
             logprob=-0.31725305,
             top_logprobs="invalid_string",
         )
+
+
+def test_delta_maps_reasoning_to_reasoning_content():
+    """
+    Test that Delta maps 'reasoning' field to 'reasoning_content'.
+
+    Providers like Cerebras and Groq return delta.reasoning for gpt-oss models,
+    but LiteLLM expects delta.reasoning_content.
+    """
+    from litellm.types.utils import Delta
+
+    # When provider sends 'reasoning' (e.g., Cerebras gpt-oss streaming)
+    delta = Delta(content=None, role="assistant", reasoning="thinking step by step")
+    assert delta.reasoning_content == "thinking step by step"
+    assert not hasattr(delta, "reasoning"), "reasoning should not leak as an extra attribute"
+
+    # When provider sends 'reasoning_content' directly (e.g., NIM), it still works
+    delta2 = Delta(content="hello", reasoning_content="direct reasoning")
+    assert delta2.reasoning_content == "direct reasoning"
+
+    # When both are present, reasoning_content takes precedence
+    delta3 = Delta(reasoning_content="from_rc", reasoning="from_r")
+    assert delta3.reasoning_content == "from_rc"
+
+    # When neither is present, reasoning_content is not set (OpenAI spec)
+    delta4 = Delta(content="hello")
+    assert not hasattr(delta4, "reasoning_content")