Merge pull request #14026 from TomeHirata/citation-supported-text

Add supported text field to anthropic citation response

Author: krrishdholakia

litellm/llms/anthropic/chat/handler.py

@@ -743,6 +743,8 @@ def chunk_parser(self, chunk: dict) -> ModelResponseStream:
                 )
 
             text, tool_use = self._handle_json_mode_chunk(text=text, tool_use=tool_use)
+            if type_chunk:
+                provider_specific_fields["chunk_type"] = type_chunk
 
             returned_chunk = ModelResponseStream(
                 choices=[

litellm/llms/anthropic/chat/transformation.py

@@ -797,7 +797,15 @@ def extract_response_content(self, completion_response: dict) -> Tuple[
             if content.get("citations") is not None:
                 if citations is None:
                     citations = []
-                citations.append(content["citations"])
+                citations.append(
+                    [
+                        {
+                            **citation,
+                            "supported_text": content.get("text", ""),
+                        }
+                        for citation in content["citations"]
+                    ]
+                )
         if thinking_blocks is not None:
             reasoning_content = ""
             for block in thinking_blocks:

tests/llm_translation/test_anthropic_completion.py

@@ -920,6 +920,14 @@ def test_anthropic_citations_api():
     citations = resp.choices[0].message.provider_specific_fields["citations"]
 
     assert citations is not None
+    if citations:
+        citation = citations[0][0]
+        assert "supported_text" in citation
+        assert "cited_text" in citation
+        assert "document_index" in citation
+        assert "document_title" in citation
+        assert "start_char_index" in citation
+        assert "end_char_index" in citation
 
 
 def test_anthropic_citations_api_streaming():
@@ -955,11 +963,11 @@ def test_anthropic_citations_api_streaming():
     has_citations = False
     for chunk in resp:
         print(f"returned chunk: {chunk}")
-        if (
-            chunk.choices[0].delta.provider_specific_fields
-            and "citation" in chunk.choices[0].delta.provider_specific_fields
-        ):
-            has_citations = True
+        if provider_specific_fields := chunk.choices[0].delta.provider_specific_fields:
+            if "citation" in provider_specific_fields:
+                has_citations = True
+
+            assert "chunk_type" in provider_specific_fields
 
     assert has_citations