include

langchain-ai · Nov 6, 2023 · b7b92d5 · b7b92d5
1 parent 8e0cb2e
commit b7b92d5
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 2 deletions.
diff --git a/libs/langchain/langchain/chat_models/openai.py b/libs/langchain/langchain/chat_models/openai.py
@@ -342,6 +342,7 @@ def _completion_with_retry(**kwargs: Any) -> Any:
 
     def _combine_llm_outputs(self, llm_outputs: List[Optional[dict]]) -> dict:
         overall_token_usage: dict = {}
+        system_fingerprint = None
         for output in llm_outputs:
             if output is None:
                 # Happens in streaming
@@ -352,7 +353,12 @@ def _combine_llm_outputs(self, llm_outputs: List[Optional[dict]]) -> dict:
                     overall_token_usage[k] += v
                 else:
                     overall_token_usage[k] = v
-        return {"token_usage": overall_token_usage, "model_name": self.model_name}
+            if system_fingerprint is None:
+                system_fingerprint = output.get("system_fingerprint")
+        combined = {"token_usage": overall_token_usage, "model_name": self.model_name}
+        if system_fingerprint:
+            combined["system_fingerprint"] = system_fingerprint
+        return combined
 
     def _stream(
         self,
@@ -430,7 +436,11 @@ def _create_chat_result(self, response: Union[dict, BaseModel]) -> ChatResult:
             )
             generations.append(gen)
         token_usage = response.get("usage", {})
-        llm_output = {"token_usage": token_usage, "model_name": self.model_name}
+        llm_output = {
+            "token_usage": token_usage,
+            "model_name": self.model_name,
+            "system_fingerprint": response.get("system_fingerprint", ""),
+        }
         return ChatResult(generations=generations, llm_output=llm_output)
 
     async def _astream(

diff --git a/libs/langchain/tests/integration_tests/chat_models/test_openai.py b/libs/langchain/tests/integration_tests/chat_models/test_openai.py
@@ -58,6 +58,7 @@ def test_chat_openai_generate() -> None:
     response = chat.generate([[message], [message]])
     assert isinstance(response, LLMResult)
     assert len(response.generations) == 2
+    assert "system_fingerprint" in response.llm_output
     for generations in response.generations:
         assert len(generations) == 2
         for generation in generations:
@@ -163,6 +164,7 @@ async def test_async_chat_openai() -> None:
     response = await chat.agenerate([[message], [message]])
     assert isinstance(response, LLMResult)
     assert len(response.generations) == 2
+    assert "system_fingerprint" in response.llm_output
     for generations in response.generations:
         assert len(generations) == 2
         for generation in generations: