diff --git a/litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py b/litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py
index d7b527824ea..1d86c105e61 100644
--- a/litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py
+++ b/litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py
@@ -455,6 +455,14 @@ class LiteLLMAnthropicMessagesAdapter:
                                             )
                                             self._add_cache_control_if_applicable(content, tool_result, model)
                                             tool_message_list.append(tool_result)
+                                        elif c.get("type") == "tool_reference":
+                                            tool_result = ChatCompletionToolMessage(
+                                                role="tool",
+                                                tool_call_id=content.get("tool_use_id", ""),
+                                                content=f"[Loaded tool: {c.get('tool_name', '')}]",
+                                            )
+                                            self._add_cache_control_if_applicable(content, tool_result, model)
+                                            tool_message_list.append(tool_result)
                                         elif c.get("type") in ("image", "document"):
                                             image_part = self._tool_result_image_part(c.get("source"))
                                             tool_result = ChatCompletionToolMessage(
@@ -483,6 +491,13 @@ class LiteLLMAnthropicMessagesAdapter:
                                                         text=c.get("text", ""),
                                                     )
                                                 )
+                                            elif c.get("type") == "tool_reference":
+                                                combined_content_parts.append(
+                                                    ChatCompletionTextObject(
+                                                        type="text",
+                                                        text=f"[Loaded tool: {c.get('tool_name', '')}]",
+                                                    )
+                                                )
                                             elif c.get("type") in ("image", "document"):
                                                 image_part = self._tool_result_image_part(c.get("source"))
                                                 if image_part:
diff --git a/litellm/llms/anthropic/experimental_pass_through/responses_adapters/transformation.py b/litellm/llms/anthropic/experimental_pass_through/responses_adapters/transformation.py
index ace7fc25dc9..f392c5335dd 100644
--- a/litellm/llms/anthropic/experimental_pass_through/responses_adapters/transformation.py
+++ b/litellm/llms/anthropic/experimental_pass_through/responses_adapters/transformation.py
@@ -294,7 +294,11 @@ class LiteLLMAnthropicToResponsesAPIAdapter:
                                 output_text = inner
                             elif isinstance(inner, list):
                                 parts = [
-                                    c.get("text", "") for c in inner if isinstance(c, dict) and c.get("type") == "text"
+                                    c.get("text", "")
+                                    if c.get("type") == "text"
+                                    else f"[Loaded tool: {c.get('tool_name', '')}]"
+                                    for c in inner
+                                    if isinstance(c, dict) and c.get("type") in {"text", "tool_reference"}
                                 ]
                                 output_text = "\n".join(parts)
                                 image_candidates = tuple(
diff --git a/litellm/types/llms/anthropic.py b/litellm/types/llms/anthropic.py
index 901802a6640..7805dd595a2 100644
--- a/litellm/types/llms/anthropic.py
+++ b/litellm/types/llms/anthropic.py
@@ -324,7 +324,12 @@ class AnthropicMessagesToolResultParam(TypedDict, total=False):
     is_error: bool
     content: (
         str
-        | Iterable[AnthropicMessagesToolResultContent | AnthropicMessagesImageParam | AnthropicMessagesDocumentParam]
+        | Iterable[
+            AnthropicMessagesToolResultContent
+            | AnthropicMessagesImageParam
+            | AnthropicMessagesDocumentParam
+            | ToolReference
+        ]
     )
     cache_control: dict | ChatCompletionCachedContent | None
 
diff --git a/tests/test_litellm/llms/anthropic/experimental_pass_through/adapters/test_anthropic_experimental_pass_through_adapters_transformation.py b/tests/test_litellm/llms/anthropic/experimental_pass_through/adapters/test_anthropic_experimental_pass_through_adapters_transformation.py
index 95fbd06547b..8bc865181df 100644
--- a/tests/test_litellm/llms/anthropic/experimental_pass_through/adapters/test_anthropic_experimental_pass_through_adapters_transformation.py
+++ b/tests/test_litellm/llms/anthropic/experimental_pass_through/adapters/test_anthropic_experimental_pass_through_adapters_transformation.py
@@ -1610,6 +1610,70 @@ def test_translate_anthropic_messages_to_openai_tool_result_single_item_backward
     assert tool_message["content"] == "72°F and sunny"
 
 
+@pytest.mark.parametrize(
+    ("tool_references", "expected_content"),
+    [
+        (
+            [{"type": "tool_reference", "tool_name": "get_weather"}],
+            "[Loaded tool: get_weather]",
+        ),
+        (
+            [
+                {"type": "tool_reference", "tool_name": "get_weather"},
+                {"type": "tool_reference", "tool_name": "search_docs"},
+            ],
+            [
+                {"type": "text", "text": "[Loaded tool: get_weather]"},
+                {"type": "text", "text": "[Loaded tool: search_docs]"},
+            ],
+        ),
+    ],
+)
+def test_translate_anthropic_messages_to_openai_tool_reference_results(
+    tool_references: list[dict[str, str]], expected_content: Any
+):
+    anthropic_messages = [
+        AnthopicMessagesAssistantMessageParam(
+            role="assistant",
+            content=[
+                {
+                    "type": "tool_use",
+                    "id": "toolu_tool_search",
+                    "name": "ToolSearch",
+                    "input": {"query": "weather"},
+                }
+            ],
+        ),
+        AnthropicMessagesUserMessageParam(
+            role="user",
+            content=[
+                {
+                    "type": "tool_result",
+                    "tool_use_id": "toolu_tool_search",
+                    "content": tool_references,
+                }
+            ],
+        ),
+    ]
+
+    result = LiteLLMAnthropicMessagesAdapter().translate_anthropic_messages_to_openai(
+        messages=anthropic_messages
+    )
+
+    tool_messages = [
+        message
+        for message in result
+        if isinstance(message, dict) and message.get("role") == "tool"
+    ]
+    assert tool_messages == [
+        {
+            "role": "tool",
+            "tool_call_id": "toolu_tool_search",
+            "content": expected_content,
+        }
+    ]
+
+
 def test_streaming_chunk_with_both_text_and_tool_calls_issue_18238():
     """
     When a streaming choice contains both text content and tool_calls,
diff --git a/tests/test_litellm/llms/anthropic/experimental_pass_through/responses_adapters/test_responses_adapters_transformation.py b/tests/test_litellm/llms/anthropic/experimental_pass_through/responses_adapters/test_responses_adapters_transformation.py
index 5ecf604f096..110146e2347 100644
--- a/tests/test_litellm/llms/anthropic/experimental_pass_through/responses_adapters/test_responses_adapters_transformation.py
+++ b/tests/test_litellm/llms/anthropic/experimental_pass_through/responses_adapters/test_responses_adapters_transformation.py
@@ -482,6 +482,33 @@ class TestTranslateMessagesToResponsesInput:
         result = _translate_messages(messages)
         assert result[0]["output"] == "Line 1\nLine 2"
 
+    def test_user_tool_result_list_content_preserves_tool_references(self):
+        messages = [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "tool_result",
+                        "tool_use_id": "call_search",
+                        "content": [
+                            {"type": "text", "text": "Available:"},
+                            {"type": "tool_reference", "tool_name": "get_weather"},
+                        ],
+                    }
+                ],
+            }
+        ]
+
+        result = _translate_messages(messages)
+
+        assert result == [
+            {
+                "type": "function_call_output",
+                "call_id": "call_search",
+                "output": "Available:\n[Loaded tool: get_weather]",
+            }
+        ]
+
     def test_user_tool_result_null_content(self):
         """tool_result with null content becomes empty string output."""
         messages = [