Implement a router for openai functions (#8589)

libs/langchain/langchain/chains/base.py

@@ -62,7 +62,14 @@ class Chain(Serializable, Runnable[Dict[str, Any], Dict[str, Any]], ABC):
         config: Optional[RunnableConfig] = None,
         **kwargs: Any,
     ) -> Dict[str, Any]:
-        return self(input, **(config or {}), **kwargs)
+        config = config or {}
+        return self(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
+        )
 
     async def ainvoke(
         self,
@@ -76,7 +83,14 @@ class Chain(Serializable, Runnable[Dict[str, Any], Dict[str, Any]], ABC):
                 None, partial(self.invoke, input, config, **kwargs)
             )
 
-        return await self.acall(input, **(config or {}), **kwargs)
+        config = config or {}
+        return await self.acall(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
+        )
 
     memory: Optional[BaseMemory] = None
     """Optional memory object. Defaults to None.

libs/langchain/langchain/chat_models/base.py

@@ -103,12 +103,18 @@ class BaseChatModel(BaseLanguageModel[BaseMessageChunk], ABC):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> BaseMessageChunk:
+        config = config or {}
         return cast(
             BaseMessageChunk,
             cast(
                 ChatGeneration,
                 self.generate_prompt(
-                    [self._convert_input(input)], stop=stop, **(config or {}), **kwargs
+                    [self._convert_input(input)],
+                    stop=stop,
+                    callbacks=config.get("callbacks"),
+                    tags=config.get("tags"),
+                    metadata=config.get("metadata"),
+                    **kwargs,
                 ).generations[0][0],
             ).message,
         )
@@ -127,8 +133,14 @@ class BaseChatModel(BaseLanguageModel[BaseMessageChunk], ABC):
                 None, partial(self.invoke, input, config, stop=stop, **kwargs)
             )
 
+        config = config or {}
         llm_result = await self.agenerate_prompt(
-            [self._convert_input(input)], stop=stop, **(config or {}), **kwargs
+            [self._convert_input(input)],
+            stop=stop,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
         )
         return cast(
             BaseMessageChunk, cast(ChatGeneration, llm_result.generations[0][0]).message

libs/langchain/langchain/llms/base.py

@@ -219,9 +219,15 @@ class BaseLLM(BaseLanguageModel[str], ABC):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> str:
+        config = config or {}
         return (
             self.generate_prompt(
-                [self._convert_input(input)], stop=stop, **(config or {}), **kwargs
+                [self._convert_input(input)],
+                stop=stop,
+                callbacks=config.get("callbacks"),
+                tags=config.get("tags"),
+                metadata=config.get("metadata"),
+                **kwargs,
             )
             .generations[0][0]
             .text
@@ -241,8 +247,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
                 None, partial(self.invoke, input, config, stop=stop, **kwargs)
             )
 
+        config = config or {}
         llm_result = await self.agenerate_prompt(
-            [self._convert_input(input)], stop=stop, **(config or {}), **kwargs
+            [self._convert_input(input)],
+            stop=stop,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
         )
         return llm_result.generations[0][0].text
 

libs/langchain/langchain/runnables/openai_functions.py

@@ -0,0 +1,46 @@
+from operator import itemgetter
+from typing import Any, Callable, List, Mapping, Optional, Union
+
+from typing_extensions import TypedDict
+
+from langchain.output_parsers.openai_functions import JsonOutputFunctionsParser
+from langchain.schema.output import ChatGeneration
+from langchain.schema.runnable import RouterRunnable, Runnable, RunnableBinding
+
+
+class OpenAIFunction(TypedDict):
+    """A function description for ChatOpenAI"""
+
+    name: str
+    """The name of the function."""
+    description: str
+    """The description of the function."""
+    parameters: dict
+    """The parameters to the function."""
+
+
+class OpenAIFunctionsRouter(RunnableBinding[ChatGeneration, Any]):
+    """A runnable that routes to the selected function."""
+
+    functions: Optional[List[OpenAIFunction]]
+
+    def __init__(
+        self,
+        runnables: Mapping[
+            str,
+            Union[
+                Runnable[dict, Any],
+                Callable[[dict], Any],
+            ],
+        ],
+        functions: Optional[List[OpenAIFunction]] = None,
+    ):
+        if functions is not None:
+            assert len(functions) == len(runnables)
+            assert all(func["name"] in runnables for func in functions)
+        router = (
+            JsonOutputFunctionsParser(args_only=False)
+            | {"key": itemgetter("name"), "input": itemgetter("arguments")}
+            | RouterRunnable(runnables)
+        )
+        super().__init__(bound=router, kwargs={}, functions=functions)

libs/langchain/langchain/schema/retriever.py

@@ -107,7 +107,13 @@ class BaseRetriever(Serializable, Runnable[str, List[Document]], ABC):
     def invoke(
         self, input: str, config: Optional[RunnableConfig] = None
     ) -> List[Document]:
-        return self.get_relevant_documents(input, **(config or {}))
+        config = config or {}
+        return self.get_relevant_documents(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+        )
 
     async def ainvoke(
         self, input: str, config: Optional[RunnableConfig] = None
@@ -116,7 +122,13 @@ class BaseRetriever(Serializable, Runnable[str, List[Document]], ABC):
             # If the retriever doesn't implement async, use default implementation
             return await super().ainvoke(input, config)
 
-        return await self.aget_relevant_documents(input, **(config or {}))
+        config = config or {}
+        return await self.aget_relevant_documents(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+        )
 
     @abstractmethod
     def _get_relevant_documents(

libs/langchain/langchain/schema/runnable.py

@@ -1254,7 +1254,7 @@ class RunnablePassthrough(Serializable, Runnable[Input, Input]):
 
 class RunnableBinding(Serializable, Runnable[Input, Output]):
     """
-    A runnable that binds a runnable to a set of kwargs.
+    A runnable that delegates calls to another runnable with a set of kwargs.
     """
 
     bound: Runnable[Input, Output]
@@ -1339,8 +1339,15 @@ class RouterRunnable(
 
     runnables: Mapping[str, Runnable[Input, Output]]
 
-    def __init__(self, runnables: Mapping[str, Runnable[Input, Output]]) -> None:
-        super().__init__(runnables=runnables)
+    def __init__(
+        self,
+        runnables: Mapping[
+            str, Union[Runnable[Input, Output], Callable[[Input], Output]]
+        ],
+    ) -> None:
+        super().__init__(
+            runnables={key: _coerce_to_runnable(r) for key, r in runnables.items()}
+        )
 
     class Config:
         arbitrary_types_allowed = True

libs/langchain/langchain/tools/base.py

@@ -203,7 +203,13 @@ class BaseTool(BaseModel, Runnable[Union[str, Dict], Any], metaclass=ToolMetacla
         **kwargs: Any,
     ) -> Any:
         config = config or {}
-        return self.run(input, **config, **kwargs)
+        return self.run(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
+        )
 
     async def ainvoke(
         self,
@@ -216,7 +222,13 @@ class BaseTool(BaseModel, Runnable[Union[str, Dict], Any], metaclass=ToolMetacla
             return super().ainvoke(input, config, **kwargs)
 
         config = config or {}
-        return await self.arun(input, **config, **kwargs)
+        return await self.arun(
+            input,
+            callbacks=config.get("callbacks"),
+            tags=config.get("tags"),
+            metadata=config.get("metadata"),
+            **kwargs,
+        )
 
     # --- Tool ---
 

libs/langchain/tests/unit_tests/runnables/__snapshots__/test_openai_functions.ambr

@@ -0,0 +1,31 @@
+# serializer version: 1
+# name: test_openai_functions_router
+  list([
+    dict({
+      'description': 'Sends the draft for revision.',
+      'name': 'revise',
+      'parameters': dict({
+        'properties': dict({
+          'notes': dict({
+            'description': "The editor's notes to guide the revision.",
+            'type': 'string',
+          }),
+        }),
+        'type': 'object',
+      }),
+    }),
+    dict({
+      'description': 'Accepts the draft.',
+      'name': 'accept',
+      'parameters': dict({
+        'properties': dict({
+          'draft': dict({
+            'description': 'The draft to accept.',
+            'type': 'string',
+          }),
+        }),
+        'type': 'object',
+      }),
+    }),
+  ])
+# ---

libs/langchain/tests/unit_tests/runnables/test_openai_functions.py

@@ -0,0 +1,95 @@
+from typing import Any, List, Optional
+
+from pytest_mock import MockerFixture
+from syrupy import SnapshotAssertion
+
+from langchain.callbacks.manager import CallbackManagerForLLMRun
+from langchain.chat_models.base import BaseChatModel
+from langchain.runnables.openai_functions import OpenAIFunctionsRouter
+from langchain.schema import ChatResult
+from langchain.schema.messages import AIMessage, BaseMessage
+from langchain.schema.output import ChatGeneration
+
+
+class FakeChatOpenAI(BaseChatModel):
+    @property
+    def _llm_type(self) -> str:
+        return "fake-openai-chat-model"
+
+    def _generate(
+        self,
+        messages: List[BaseMessage],
+        stop: Optional[List[str]] = None,
+        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        **kwargs: Any,
+    ) -> ChatResult:
+        return ChatResult(
+            generations=[
+                ChatGeneration(
+                    message=AIMessage(
+                        content="",
+                        additional_kwargs={
+                            "function_call": {
+                                "name": "accept",
+                                "arguments": '{\n  "draft": "turtles"\n}',
+                            }
+                        },
+                    )
+                )
+            ]
+        )
+
+
+def test_openai_functions_router(
+    snapshot: SnapshotAssertion, mocker: MockerFixture
+) -> None:
+    revise = mocker.Mock(
+        side_effect=lambda kw: f'Revised draft: no more {kw["notes"]}!'
+    )
+    accept = mocker.Mock(side_effect=lambda kw: f'Accepted draft: {kw["draft"]}!')
+
+    router = OpenAIFunctionsRouter(
+        {
+            "revise": revise,
+            "accept": accept,
+        },
+        functions=[
+            {
+                "name": "revise",
+                "description": "Sends the draft for revision.",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "notes": {
+                            "type": "string",
+                            "description": "The editor's notes to guide the revision.",
+                        },
+                    },
+                },
+            },
+            {
+                "name": "accept",
+                "description": "Accepts the draft.",
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "draft": {
+                            "type": "string",
+                            "description": "The draft to accept.",
+                        },
+                    },
+                },
+            },
+        ],
+    )
+
+    model = FakeChatOpenAI()
+
+    chain = model.bind(functions=router.functions) | router
+
+    assert router.functions == snapshot
+
+    assert chain.invoke("Something about turtles?") == "Accepted draft: turtles!"
+
+    revise.assert_not_called()
+    accept.assert_called_once_with({"draft": "turtles"})