Update docs links

.clabot

@@ -1,4 +1,4 @@
 {
-  "contributors": ["eyurtsev", "hwchase17", "nfcampos", "efriis", "jacoblee93", "dqbd", "kreneskyp", "adarsh-jha-dev", "harris", "baskaryan", "hinthornw", "bracesproul", "jakerachleff", "craigsdennis", "anhi", "169", "LarchLiu", "PaulLockett", "RCMatthias", "jwynia", "majiayu000", "mpskex", "shivachittamuru", "sinashaloudegi", "sowsan", "akira", "lucianotonet", "JGalego", "nat-n", "dirien", "donbr", "rahilvora", "WarrenTheRabbit", "StreetLamb", "ccurme", "dennisrall"],
+  "contributors": ["eyurtsev", "hwchase17", "nfcampos", "efriis", "jacoblee93", "dqbd", "kreneskyp", "adarsh-jha-dev", "harris", "baskaryan", "hinthornw", "bracesproul", "jakerachleff", "craigsdennis", "anhi", "169", "LarchLiu", "PaulLockett", "RCMatthias", "jwynia", "majiayu000", "mpskex", "shivachittamuru", "sinashaloudegi", "sowsan", "akira", "lucianotonet", "JGalego", "nat-n", "dirien", "donbr", "rahilvora", "WarrenTheRabbit", "StreetLamb", "ccurme"],
   "message": "Thank you for your pull request and welcome to our community. We require contributors to sign our Contributor License Agreement, and we don't seem to have the username {{usersWithoutCLA}} on file. In order for us to review and merge your code, please complete the Individual Contributor License Agreement here https://forms.gle/AQFbtkWRoHXUgipM6 .\n\nThis process is done manually on our side, so after signing the form one of the maintainers will add you to the contributors list.\n\nFor more details about why we have a CLA and other contribution guidelines please see: https://github.com/langchain-ai/langserve/blob/main/CONTRIBUTING.md."
 }

.github/workflows/_pydantic_compatibility.yml

@@ -0,0 +1,94 @@
+name: pydantic v1/v2 compatibility
+
+on:
+  workflow_call:
+    inputs:
+      working-directory:
+        required: true
+        type: string
+        description: "From which folder this pipeline executes"
+
+env:
+  POETRY_VERSION: "1.5.1"
+
+jobs:
+  build:
+    timeout-minutes: 10
+    defaults:
+      run:
+        working-directory: ${{ inputs.working-directory }}
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version:
+          - "3.8"
+          - "3.9"
+          - "3.10"
+          - "3.11"
+    name: Pydantic v1/v2 compatibility - Python ${{ matrix.python-version }}
+    steps:
+      - uses: actions/checkout@v3
+
+      - name: Set up Python ${{ matrix.python-version }} + Poetry ${{ env.POETRY_VERSION }}
+        uses: "./.github/actions/poetry_setup"
+        with:
+          python-version: ${{ matrix.python-version }}
+          poetry-version: ${{ env.POETRY_VERSION }}
+          working-directory: ${{ inputs.working-directory }}
+          cache-key: pydantic-cross-compat
+
+      - name: Install dependencies
+        shell: bash
+        run: poetry install
+
+      - name: Install the opposite major version of pydantic
+        # If normal tests use pydantic v1, here we'll use v2, and vice versa.
+        shell: bash
+        run: |
+          # Determine the major part of pydantic version
+          REGULAR_VERSION=$(poetry run python -c "import pydantic; print(pydantic.__version__)" | cut -d. -f1)
+
+          if [[ "$REGULAR_VERSION" == "1" ]]; then
+            PYDANTIC_DEP=">=2.1,<3"
+            TEST_WITH_VERSION="2"
+          elif [[ "$REGULAR_VERSION" == "2" ]]; then
+            PYDANTIC_DEP="<2"
+            TEST_WITH_VERSION="1"
+          else
+            echo "Unexpected pydantic major version '$REGULAR_VERSION', cannot determine which version to use for cross-compatibility test."
+            exit 1
+          fi
+
+          # Install via `pip` instead of `poetry add` to avoid changing lockfile,
+          # which would prevent caching from working: the cache would get saved
+          # to a different key than where it gets loaded from.
+          poetry run pip install "pydantic${PYDANTIC_DEP}"
+
+          # Ensure that the correct pydantic is installed now.
+          echo "Checking pydantic version... Expecting ${TEST_WITH_VERSION}"
+
+          # Determine the major part of pydantic version
+          CURRENT_VERSION=$(poetry run python -c "import pydantic; print(pydantic.__version__)" | cut -d. -f1)
+
+          # Check that the major part of pydantic version is as expected, if not
+          # raise an error
+          if [[ "$CURRENT_VERSION" != "$TEST_WITH_VERSION" ]]; then
+            echo "Error: expected pydantic version ${CURRENT_VERSION} to have been installed, but found: ${TEST_WITH_VERSION}"
+            exit 1
+          fi
+          echo "Found pydantic version ${CURRENT_VERSION}, as expected"
+      - name: Run pydantic compatibility tests
+        shell: bash
+        run: make test
+
+      - name: Ensure the tests did not create any additional files
+        shell: bash
+        run: |
+          set -eu
+
+          STATUS="$(git status)"
+          echo "$STATUS"
+
+          # grep will exit non-zero if the target message isn't found,
+          # and `set -e` above will cause the step to fail.
+          echo "$STATUS" | grep 'nothing to commit, working tree clean'

.github/workflows/langserve_ci.yml

@@ -39,6 +39,13 @@ jobs:
     with:
       working-directory: .
     secrets: inherit
+
+  pydantic-compatibility:
+    uses:
+      ./.github/workflows/_pydantic_compatibility.yml
+    with:
+      working-directory: .
+    secrets: inherit
   test:
     timeout-minutes: 10
     runs-on: ubuntu-latest

.github/workflows/langserve_release.yml

@@ -10,5 +10,4 @@ jobs:
       ./.github/workflows/_release.yml
     with:
       working-directory: .
-    permissions: write-all
     secrets: inherit

README.md

@@ -5,10 +5,14 @@
 [![Open Issues](https://img.shields.io/github/issues-raw/langchain-ai/langserve)](https://github.com/langchain-ai/langserve/issues)
 [![](https://dcbadge.vercel.app/api/server/6adMQxSpJS?compact=true&style=flat)](https://discord.com/channels/1038097195422978059/1170024642245832774)
 
+🚩 We will be releasing a hosted version of LangServe for one-click deployments of
+LangChain applications. [Sign up here](https://forms.gle/KC13Nzn76UeLaghK7)
+to get on the waitlist.
+
 ## Overview
 
 [LangServe](https://github.com/langchain-ai/langserve) helps developers
-deploy `LangChain` [runnables and chains](https://python.langchain.com/docs/expression_language/)
+deploy `LangChain` [runnables and chains](https://python.langchain.com/v0.2/docs/how_to/#langchain-expression-language-lcel)
 as a REST API.
 
 This library is integrated with [FastAPI](https://fastapi.tiangolo.com/) and
@@ -38,13 +42,6 @@ in [LangChain.js](https://js.langchain.com/docs/ecosystem/langserve).
   locally (or call the HTTP API directly)
 - [LangServe Hub](https://github.com/langchain-ai/langchain/blob/master/templates/README.md)
 
-## ⚠️ LangGraph Compatibility
-
-LangServe is designed to primarily deploy simple Runnables and wok with well-known primitives in langchain-core.
-
-If you need a deployment option for LangGraph, you should instead be looking at [LangGraph Cloud (beta)](https://langchain-ai.github.io/langgraph/cloud/) which will
-be better suited for deploying LangGraph applications.
-
 ## Limitations
 
 - Client callbacks are not yet supported for events that originate on the server
@@ -52,6 +49,13 @@ be better suited for deploying LangGraph applications.
   support [mixing pydantic v1 and v2 namespaces](https://github.com/tiangolo/fastapi/issues/10360).
   See section below for more details.
 
+## Hosted LangServe
+
+We will be releasing a hosted version of LangServe for one-click deployments of
+LangChain
+applications. [Sign up here](https://forms.gle/KC13Nzn76UeLaghK7)
+to get on the waitlist.
+
 ## Security
 
 - Vulnerability in Versions 0.0.13 - 0.0.15 -- playground endpoint allows accessing
@@ -124,13 +128,13 @@ directory.
 | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | **LLMs** Minimal example that reserves OpenAI and Anthropic chat models. Uses async, supports batching and streaming.                                                                                                                                              | [server](https://github.com/langchain-ai/langserve/tree/main/examples/llm/server.py), [client](https://github.com/langchain-ai/langserve/blob/main/examples/llm/client.ipynb)                                                       |
 | **Retriever** Simple server that exposes a retriever as a runnable.                                                                                                                                                                                                | [server](https://github.com/langchain-ai/langserve/tree/main/examples/retrieval/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/retrieval/client.ipynb)                                           |
-| **Conversational Retriever** A [Conversational Retriever](https://python.langchain.com/docs/expression_language/cookbook/retrieval#conversational-retrieval-chain) exposed via LangServe                                                                           | [server](https://github.com/langchain-ai/langserve/tree/main/examples/conversational_retrieval_chain/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/conversational_retrieval_chain/client.ipynb) |
+| **Conversational Retriever** A Conversational Retriever exposed via LangServe                                                                           | [server](https://github.com/langchain-ai/langserve/tree/main/examples/conversational_retrieval_chain/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/conversational_retrieval_chain/client.ipynb) |
 | **Agent** without **conversation history** based on [OpenAI tools](https://python.langchain.com/docs/modules/agents/agent_types/openai_functions_agent)                                                                                                            | [server](https://github.com/langchain-ai/langserve/tree/main/examples/agent/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/agent/client.ipynb)                                                   |
 | **Agent** with **conversation history** based on [OpenAI tools](https://python.langchain.com/docs/modules/agents/agent_types/openai_functions_agent)                                                                                                               | [server](https://github.com/langchain-ai/langserve/blob/main/examples/agent_with_history/server.py), [client](https://github.com/langchain-ai/langserve/blob/main/examples/agent_with_history/client.ipynb)                         |
-| [RunnableWithMessageHistory](https://python.langchain.com/docs/expression_language/how_to/message_history) to implement chat persisted on backend, keyed off a `session_id` supplied by client.                                                                    | [server](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence/client.ipynb)                   |
-| [RunnableWithMessageHistory](https://python.langchain.com/docs/expression_language/how_to/message_history) to implement chat persisted on backend, keyed off a `conversation_id` supplied by client, and `user_id` (see Auth for implementing `user_id` properly). | [server](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence_and_user/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence_and_user/client.ipynb) |
-| [Configurable Runnable](https://python.langchain.com/docs/expression_language/how_to/configure) to create a retriever that supports run time configuration of the index name.                                                                                      | [server](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_retrieval/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_retrieval/client.ipynb)                 |
-| [Configurable Runnable](https://python.langchain.com/docs/expression_language/how_to/configure) that shows configurable fields and configurable alternatives.                                                                                                      | [server](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_chain/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_chain/client.ipynb)                         |
+| [RunnableWithMessageHistory](https://python.langchain.com/v0.2/docs/how_to/message_history/) to implement chat persisted on backend, keyed off a `session_id` supplied by client.                                                                    | [server](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence/client.ipynb)                   |
+| [RunnableWithMessageHistory](https://python.langchain.com/v0.2/docs/how_to/message_history/) to implement chat persisted on backend, keyed off a `conversation_id` supplied by client, and `user_id` (see Auth for implementing `user_id` properly). | [server](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence_and_user/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/chat_with_persistence_and_user/client.ipynb) |
+| [Configurable Runnable](https://python.langchain.com/v0.2/docs/how_to/configure/) to create a retriever that supports run time configuration of the index name.                                                                                      | [server](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_retrieval/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_retrieval/client.ipynb)                 |
+| [Configurable Runnable](https://python.langchain.com/v0.2/docs/how_to/configure/) that shows configurable fields and configurable alternatives.                                                                                                      | [server](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_chain/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/configurable_chain/client.ipynb)                         |
 | **APIHandler** Shows how to use `APIHandler` instead of `add_routes`. This provides more flexibility for developers to define endpoints. Works well with all FastAPI patterns, but takes a bit more effort.                                                        | [server](https://github.com/langchain-ai/langserve/tree/main/examples/api_handler_examples/server.py)                                                                                                                               |
 | **LCEL Example** Example that uses LCEL to manipulate a dictionary input.                                                                                                                                                                                          | [server](https://github.com/langchain-ai/langserve/tree/main/examples/passthrough_dict/server.py), [client](https://github.com/langchain-ai/langserve/tree/main/examples/passthrough_dict/client.ipynb)                             |
 | **Auth** with `add_routes`: Simple authentication that can be applied across all endpoints associated with app. (Not useful on its own for implementing per user logic.)                                                                                           | [server](https://github.com/langchain-ai/langserve/tree/main/examples/auth/global_deps/server.py)                                                                                                                                   |
@@ -327,14 +331,14 @@ adds of these endpoints to the server:
 - `GET /my_runnable/config_schema` - json schema for config of the runnable
 
 These endpoints match
-the [LangChain Expression Language interface](https://python.langchain.com/docs/expression_language/interface) --
+the [LangChain Expression Language interface](https://python.langchain.com/v0.2/docs/how_to/lcel_cheatsheet/) --
 please reference this documentation for more details.
 
 ## Playground
 
 You can find a playground page for your runnable at `/my_runnable/playground/`. This
 exposes a simple UI
-to [configure](https://python.langchain.com/docs/expression_language/how_to/configure)
+to [configure](https://python.langchain.com/v0.2/docs/how_to/configure/)
 and invoke your runnable with streaming output and intermediate steps.
 
 <p align="center">
@@ -380,7 +384,7 @@ prompt = ChatPromptTemplate.from_messages(
     ]
 )
 
-chain = prompt | ChatAnthropic(model="claude-2.1")
+chain = prompt | ChatAnthropic(model="claude-2")
 
 
 class InputChat(BaseModel):
@@ -423,7 +427,7 @@ for demos or testing.
 ## Legacy Chains
 
 LangServe works with both Runnables (constructed
-via [LangChain Expression Language](https://python.langchain.com/docs/expression_language/))
+via [LangChain Expression Language](https://python.langchain.com/v0.2/docs/how_to/#langchain-expression-language-lcel))
 and legacy chains (inheriting from `Chain`).
 However, some of the input schemas for legacy chains may be incomplete/incorrect,
 leading to errors.
@@ -475,7 +479,7 @@ gcloud run deploy [your-service-name] --source . --port 8001 --allow-unauthentic
 LangServe provides support for Pydantic 2 with some limitations.
 
 1. OpenAPI docs will not be generated for invoke/batch/stream/stream_log when using
-   Pydantic V2. Fast API does not support [mixing pydantic v1 and v2 namespaces]. To fix this, use `pip install pydantic==1.10.17`.
+   Pydantic V2. Fast API does not support [mixing pydantic v1 and v2 namespaces].
 2. LangChain uses the v1 namespace in Pydantic v2. Please read
    the [following guidelines to ensure compatibility with LangChain](https://github.com/langchain-ai/langchain/discussions/9337)
 
@@ -772,7 +776,7 @@ prompt = ChatPromptTemplate.from_messages(
     ]
 )
 
-chain = prompt | ChatAnthropic(model="claude-2.1")
+chain = prompt | ChatAnthropic(model="claude-2")
 
 
 class MessageListInput(BaseModel):

examples/agent/server.py

@@ -23,12 +23,12 @@ from fastapi import FastAPI
 from langchain.agents import AgentExecutor
 from langchain.agents.format_scratchpad import format_to_openai_functions
 from langchain.agents.output_parsers import OpenAIFunctionsAgentOutputParser
+from langchain.pydantic_v1 import BaseModel
 from langchain_community.vectorstores import FAISS
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.tools import tool
 from langchain_core.utils.function_calling import format_tool_to_openai_function
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
-from pydantic import BaseModel
 
 from langserve import add_routes
 

examples/agent_custom_streaming/server.py

@@ -57,9 +57,9 @@ from langchain_core.runnables import RunnableLambda
 from langchain_core.tools import tool
 from langchain_core.utils.function_calling import format_tool_to_openai_tool
 from langchain_openai import ChatOpenAI
-from pydantic import BaseModel
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel
 
 prompt = ChatPromptTemplate.from_messages(
     [

examples/agent_with_history/server.py

@@ -36,9 +36,9 @@ from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.tools import tool
 from langchain_core.utils.function_calling import format_tool_to_openai_tool
 from langchain_openai import ChatOpenAI
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 prompt = ChatPromptTemplate.from_messages(
     [

examples/auth/api_handler/server.py

@@ -44,10 +44,10 @@ from langchain_core.runnables import (
 )
 from langchain_core.vectorstores import VectorStore
 from langchain_openai import OpenAIEmbeddings
-from pydantic import BaseModel, ConfigDict
 from typing_extensions import Annotated
 
 from langserve import APIHandler
+from langserve.pydantic_v1 import BaseModel
 
 
 class User(BaseModel):
@@ -150,9 +150,10 @@ class PerUserVectorstore(RunnableSerializable):
     user_id: Optional[str]
     vectorstore: VectorStore
 
-    model_config = ConfigDict(
-        arbitrary_types_allowed=True,
-    )
+    class Config:
+        # Allow arbitrary types since VectorStore is an abstract interface
+        # and not a pydantic model
+        arbitrary_types_allowed = True
 
     def _invoke(
         self, input: str, config: Optional[RunnableConfig] = None, **kwargs: Any

examples/auth/per_req_config_modifier/server.py

@@ -45,10 +45,10 @@ from langchain_core.runnables import (
 )
 from langchain_core.vectorstores import VectorStore
 from langchain_openai import OpenAIEmbeddings
-from pydantic import BaseModel, ConfigDict
 from typing_extensions import Annotated
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel
 
 
 class User(BaseModel):
@@ -147,9 +147,10 @@ class PerUserVectorstore(RunnableSerializable):
     user_id: Optional[str]
     vectorstore: VectorStore
 
-    model_config = ConfigDict(
-        arbitrary_types_allowed=True,
-    )
+    class Config:
+        # Allow arbitrary types since VectorStore is an abstract interface
+        # and not a pydantic model
+        arbitrary_types_allowed = True
 
     def _invoke(
         self, input: str, config: Optional[RunnableConfig] = None, **kwargs: Any

examples/chat_playground/legacy_input/server.py

@@ -8,9 +8,9 @@ from fastapi import FastAPI
 from langchain_anthropic import ChatAnthropic
 from langchain_core.messages import AIMessage, HumanMessage, SystemMessage
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 app = FastAPI(
     title="LangChain Server",
@@ -28,7 +28,7 @@ prompt = ChatPromptTemplate.from_messages(
     ]
 )
 
-chain = prompt | ChatAnthropic(model="claude-2.1")
+chain = prompt | ChatAnthropic(model="claude-2")
 
 
 class InputChat(BaseModel):

examples/chat_playground/server.py

@@ -8,9 +8,9 @@ from fastapi import FastAPI
 from langchain_anthropic.chat_models import ChatAnthropic
 from langchain_core.messages import AIMessage, HumanMessage, SystemMessage
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 app = FastAPI(
     title="LangChain Server",

examples/chat_with_persistence/server.py

@@ -18,9 +18,9 @@ from langchain_community.chat_message_histories import FileChatMessageHistory
 from langchain_core.chat_history import BaseChatMessageHistory
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.runnables.history import RunnableWithMessageHistory
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 
 def _is_valid_identifier(value: str) -> bool:
@@ -76,7 +76,7 @@ prompt = ChatPromptTemplate.from_messages(
     ]
 )
 
-chain = prompt | ChatAnthropic(model="claude-2.1")
+chain = prompt | ChatAnthropic(model="claude-2")
 
 
 class InputChat(BaseModel):

examples/configurable_agent_executor/server.py

@@ -20,6 +20,7 @@ from fastapi import FastAPI
 from langchain.agents import AgentExecutor
 from langchain.agents.format_scratchpad import format_to_openai_functions
 from langchain.agents.output_parsers import OpenAIFunctionsAgentOutputParser
+from langchain.pydantic_v1 import BaseModel
 from langchain_community.vectorstores import FAISS
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
 from langchain_core.runnables import (
@@ -32,7 +33,6 @@ from langchain_core.runnables.utils import Input, Output
 from langchain_core.tools import tool
 from langchain_core.utils.function_calling import format_tool_to_openai_function
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
-from pydantic import BaseModel
 
 from langserve import add_routes
 

examples/configurable_retrieval/server.py

@@ -15,9 +15,9 @@ from langchain_core.runnables import (
 )
 from langchain_core.vectorstores import VectorStore
 from langchain_openai import OpenAIEmbeddings
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 vectorstore1 = FAISS.from_texts(
     ["cats like fish", "dogs like sticks"], embedding=OpenAIEmbeddings()

examples/conversational_retrieval_chain/server.py

@@ -18,9 +18,9 @@ from langchain_core.output_parsers import StrOutputParser
 from langchain_core.prompts import ChatPromptTemplate, PromptTemplate, format_document
 from langchain_core.runnables import RunnableMap, RunnablePassthrough
 from langchain_openai import ChatOpenAI, OpenAIEmbeddings
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 _TEMPLATE = """Given the following conversation and a follow up question, rephrase the 
 follow up question to be a standalone question, in its original language.

examples/file_processing/server.py

@@ -15,10 +15,10 @@ allowing one to upload a binary file using the langserve playground UI.
 import base64
 
 from fastapi import FastAPI
+from langchain.pydantic_v1 import Field
 from langchain_community.document_loaders.parsers.pdf import PDFMinerParser
 from langchain_core.document_loaders import Blob
 from langchain_core.runnables import RunnableLambda
-from pydantic import Field
 
 from langserve import CustomUserType, add_routes
 

examples/widgets/chat/message_list/server.py

@@ -10,9 +10,9 @@ from langchain_anthropic import ChatAnthropic
 from langchain_core.messages import AIMessage, HumanMessage, SystemMessage
 from langchain_core.output_parsers import StrOutputParser
 from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
-from pydantic import BaseModel, Field
 
 from langserve import add_routes
+from langserve.pydantic_v1 import BaseModel, Field
 
 app = FastAPI(
     title="LangChain Server",
@@ -40,7 +40,7 @@ prompt = ChatPromptTemplate.from_messages(
     ]
 )
 
-chain = prompt | ChatAnthropic(model="claude-2.1") | StrOutputParser()
+chain = prompt | ChatAnthropic(model="claude-2") | StrOutputParser()
 
 
 class InputChat(BaseModel):

examples/widgets/chat/tuples/server.py

@@ -6,6 +6,7 @@ from typing import Any, Dict, List, Tuple
 
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
+from langchain.pydantic_v1 import BaseModel, Field
 from langchain_community.document_loaders.parsers.pdf import PDFMinerParser
 from langchain_core.document_loaders import Blob
 from langchain_core.messages import (
@@ -16,7 +17,6 @@ from langchain_core.messages import (
 )
 from langchain_core.runnables import RunnableLambda, RunnableParallel
 from langchain_openai import ChatOpenAI
-from pydantic import BaseModel, Field
 
 from langserve import CustomUserType
 from langserve.server import add_routes

langserve/_pydantic.py

@@ -1,53 +0,0 @@
-from typing import Any, Dict, Type, cast
-
-from pydantic import BaseModel, ConfigDict, RootModel
-from pydantic.json_schema import (
-    DEFAULT_REF_TEMPLATE,
-    GenerateJsonSchema,
-    JsonSchemaMode,
-)
-
-
-def _create_root_model(name: str, type_: Any) -> Type[RootModel]:
-    """Create a base class."""
-
-    def schema(
-        cls: Type[BaseModel],
-        by_alias: bool = True,
-        ref_template: str = DEFAULT_REF_TEMPLATE,
-    ) -> Dict[str, Any]:
-        # Complains about schema not being defined in superclass
-        schema_ = super(cls, cls).schema(  # type: ignore[misc]
-            by_alias=by_alias, ref_template=ref_template
-        )
-        schema_["title"] = name
-        return schema_
-
-    def model_json_schema(
-        cls: Type[BaseModel],
-        by_alias: bool = True,
-        ref_template: str = DEFAULT_REF_TEMPLATE,
-        schema_generator: type[GenerateJsonSchema] = GenerateJsonSchema,
-        mode: JsonSchemaMode = "validation",
-    ) -> Dict[str, Any]:
-        # Complains about model_json_schema not being defined in superclass
-        schema_ = super(cls, cls).model_json_schema(  # type: ignore[misc]
-            by_alias=by_alias,
-            ref_template=ref_template,
-            schema_generator=schema_generator,
-            mode=mode,
-        )
-        schema_["title"] = name
-        return schema_
-
-    base_class_attributes = {
-        "__annotations__": {"root": type_},
-        "model_config": ConfigDict(arbitrary_types_allowed=True),
-        "schema": classmethod(schema),
-        "model_json_schema": classmethod(model_json_schema),
-        # Should replace __module__ with caller based on stack frame.
-        "__module__": "langserve._pydantic",
-    }
-
-    custom_root_type = type(name, (RootModel,), base_class_attributes)
-    return cast(Type[RootModel], custom_root_type)

langserve/api_handler.py

@@ -29,7 +29,6 @@ from fastapi.encoders import jsonable_encoder
 from fastapi.exceptions import RequestValidationError
 from langchain_core._api.beta_decorator import warn_beta
 from langchain_core.callbacks.base import AsyncCallbackHandler
-from langchain_core.callbacks.manager import BaseCallbackManager
 from langchain_core.load.serializable import Serializable
 from langchain_core.runnables import Runnable, RunnableConfig
 from langchain_core.runnables.config import (
@@ -41,16 +40,14 @@ from langchain_core.tracers import RunLogPatch
 from langsmith import client as ls_client
 from langsmith.schemas import FeedbackIngestToken
 from langsmith.utils import tracing_is_enabled
-from pydantic import BaseModel, Field, RootModel, ValidationError, create_model
-from pydantic.v1 import BaseModel as BaseModelV1
 from starlette.requests import Request
 from starlette.responses import JSONResponse, Response
 from typing_extensions import TypedDict
 
-from langserve._pydantic import _create_root_model
 from langserve.callbacks import AsyncEventAggregatorCallback, CallbackEventDict
 from langserve.lzstring import LZString
 from langserve.playground import serve_playground
+from langserve.pydantic_v1 import BaseModel, Field, ValidationError, create_model
 from langserve.schema import (
     BatchResponseMetadata,
     CustomUserType,
@@ -62,7 +59,7 @@ from langserve.schema import (
     PublicTraceLink,
     PublicTraceLinkCreateRequest,
 )
-from langserve.serialization import Serializer, WellKnownLCSerializer
+from langserve.serialization import WellKnownLCSerializer
 from langserve.validation import (
     BatchBaseResponse,
     BatchRequestShallowValidator,
@@ -187,11 +184,11 @@ async def _unpack_request_config(
     config_dicts = []
     for config in client_sent_configs:
         if isinstance(config, str):
-            config_dicts.append(model(**_config_from_hash(config)).model_dump())
+            config_dicts.append(model(**_config_from_hash(config)).dict())
         elif isinstance(config, BaseModel):
-            config_dicts.append(config.model_dump())
+            config_dicts.append(config.dict())
         elif isinstance(config, Mapping):
-            config_dicts.append(model(**config).model_dump())
+            config_dicts.append(model(**config).dict())
         else:
             raise TypeError(f"Expected a string, dict or BaseModel got {type(config)}")
     config = merge_configs(*config_dicts)
@@ -258,12 +255,10 @@ def _update_config_with_defaults(
         }
         metadata.update(hosted_metadata)
 
-    non_overridable_default_config: RunnableConfig = {
-        "metadata": metadata,
-    }
-
-    if run_name:
-        non_overridable_default_config["run_name"] = run_name
+    non_overridable_default_config = RunnableConfig(
+        run_name=run_name,
+        metadata=metadata,
+    )
 
     # merge_configs is last-writer-wins, so we specifically pass in the
     # overridable configs first, then the user provided configs, then
@@ -284,8 +279,8 @@ def _update_config_with_defaults(
 
 def _unpack_input(validated_model: BaseModel) -> Any:
     """Unpack the decoded input from the validated model."""
-    if isinstance(validated_model, RootModel):
-        model = validated_model.root
+    if hasattr(validated_model, "__root__"):
+        model = validated_model.__root__
     else:
         model = validated_model
 
@@ -299,7 +294,7 @@ def _unpack_input(validated_model: BaseModel) -> Any:
         # This logic should be applied recursively to nested models.
         return {
             fieldname: _unpack_input(getattr(model, fieldname))
-            for fieldname in model.model_fields.keys()
+            for fieldname in model.__fields__.keys()
         }
 
     return model
@@ -309,7 +304,7 @@ def _rename_pydantic_model(model: Type[BaseModel], prefix: str) -> Type[BaseMode
     """Rename the given pydantic model to the given name."""
     return create_model(
         prefix + model.__name__,
-        __config__=model.model_config,
+        __config__=model.__config__,
         **{
             fieldname: (
                 _rename_pydantic_model(field.annotation, prefix)
@@ -318,10 +313,10 @@ def _rename_pydantic_model(model: Type[BaseModel], prefix: str) -> Type[BaseMode
                 Field(
                     field.default,
                     title=fieldname,
-                    description=field.description,
+                    description=field.field_info.description,
                 ),
             )
-            for fieldname, field in model.model_fields.items()
+            for fieldname, field in model.__fields__.items()
         },
     )
 
@@ -331,11 +326,6 @@ def _replace_non_alphanumeric_with_underscores(s: str) -> str:
     return re.sub(r"[^a-zA-Z0-9]", "_", s)
 
 
-def _schema_json(model: Type[BaseModel]) -> str:
-    """Return the JSON representation of the model schema."""
-    return json.dumps(model.model_json_schema(), sort_keys=True, indent=False)
-
-
 def _resolve_model(
     type_: Union[Type, BaseModel], default_name: str, namespace: str
 ) -> Type[BaseModel]:
@@ -343,15 +333,15 @@ def _resolve_model(
     if isclass(type_) and issubclass(type_, BaseModel):
         model = type_
     else:
-        model = _create_root_model(default_name, type_)
+        model = create_model(default_name, __root__=(type_, ...))
 
-    hash_ = _schema_json(model)
+    hash_ = model.schema_json()
 
     if model.__name__ in _SEEN_NAMES and hash_ not in _MODEL_REGISTRY:
         # If the model name has been seen before, but the model itself is different
         # generate a new name for the model.
         model_to_use = _rename_pydantic_model(model, namespace)
-        hash_ = _schema_json(model_to_use)
+        hash_ = model_to_use.schema_json()
     else:
         model_to_use = model
 
@@ -376,7 +366,11 @@ def _add_namespace_to_model(namespace: str, model: Type[BaseModel]) -> Type[Base
         A new model with name prepended with the given namespace.
     """
     model_with_unique_name = _rename_pydantic_model(model, namespace)
-    model_with_unique_name.model_rebuild()
+    if "run_id" in model_with_unique_name.__annotations__:
+        # Help resolve reference by providing namespace references
+        model_with_unique_name.update_forward_refs(uuid=uuid)
+    else:
+        model_with_unique_name.update_forward_refs()
     return model_with_unique_name
 
 
@@ -409,7 +403,7 @@ def _with_validation_error_translation() -> Generator[None, None, None]:
     try:
         yield
     except ValidationError as e:
-        raise RequestValidationError(e.errors())
+        raise RequestValidationError(e.errors(), body=e.model)
 
 
 def _json_encode_response(model: BaseModel) -> JSONResponse:
@@ -429,27 +423,27 @@ def _json_encode_response(model: BaseModel) -> JSONResponse:
 
     if isinstance(model, InvokeBaseResponse):
         # Invoke Response
-        # Collapse 'root' from output field if it exists. This is done
+        # Collapse '__root__' from output field if it exists. This is done
         # automatically by fastapi when annotating request and response with
         # We need to do this manually since we're using vanilla JSONResponse
-        if isinstance(obj["output"], dict) and "root" in obj["output"]:
-            obj["output"] = obj["output"]["root"]
+        if isinstance(obj["output"], dict) and "__root__" in obj["output"]:
+            obj["output"] = obj["output"]["__root__"]
 
         if "callback_events" in obj:
             for idx, callback_event in enumerate(obj["callback_events"]):
-                if isinstance(callback_event, dict) and "root" in callback_event:
-                    obj["callback_events"][idx] = callback_event["root"]
+                if isinstance(callback_event, dict) and "__root__" in callback_event:
+                    obj["callback_events"][idx] = callback_event["__root__"]
     elif isinstance(model, BatchBaseResponse):
         if not isinstance(obj["output"], list):
             raise AssertionError("Expected output to be a list")
 
-        # Collapse 'root' from output field if it exists. This is done
+        # Collapse '__root__' from output field if it exists. This is done
         # automatically by fastapi when annotating request and response with
         # We need to do this manually since we're using vanilla JSONResponse
         outputs = obj["output"]
         for idx, output in enumerate(outputs):
-            if isinstance(output, dict) and "root" in output:
-                outputs[idx] = output["root"]
+            if isinstance(output, dict) and "__root__" in output:
+                outputs[idx] = output["__root__"]
 
         if "callback_events" in obj:
             if not isinstance(obj["callback_events"], list):
@@ -457,8 +451,11 @@ def _json_encode_response(model: BaseModel) -> JSONResponse:
 
             for callback_events in obj["callback_events"]:
                 for idx, callback_event in enumerate(callback_events):
-                    if isinstance(callback_event, dict) and "root" in callback_event:
-                        callback_events[idx] = callback_event["root"]
+                    if (
+                        isinstance(callback_event, dict)
+                        and "__root__" in callback_event
+                    ):
+                        callback_events[idx] = callback_event["__root__"]
     else:
         raise AssertionError(
             f"Expected a InvokeBaseResponse or BatchBaseResponse got: {type(model)}"
@@ -473,12 +470,7 @@ def _add_callbacks(
     """Add the callback aggregator to the config."""
     if "callbacks" not in config:
         config["callbacks"] = []
-    if "callbacks" in config:
-        if isinstance(config["callbacks"], list):
-            config["callbacks"].extend(callbacks)
-        elif isinstance(config["callbacks"], BaseCallbackManager):
-            for callback in callbacks:
-                config["callbacks"].add_handler(callback, inherit=True)
+    config["callbacks"].extend(callbacks)
 
 
 _MODEL_REGISTRY = {}
@@ -535,8 +527,6 @@ class APIHandler:
         per_req_config_modifier: Optional[PerRequestConfigModifier] = None,
         stream_log_name_allow_list: Optional[Sequence[str]] = None,
         playground_type: Literal["default", "chat"] = "default",
-        astream_events_version: Literal["v1", "v2"] = "v2",
-        serializer: Optional[Serializer] = None,
     ) -> None:
         """Create an API handler for the given runnable.
 
@@ -571,7 +561,6 @@ class APIHandler:
                 If true, the client will be able to show trace information
                 including events that occurred on the server side.
                 Be sure not to include any sensitive information in the callback events.
-                This is a **beta** API.
             enable_feedback_endpoint: Whether to enable an endpoint for logging feedback
                 to LangSmith. Disabled by default. If this flag is disabled or LangSmith
                 tracing is not enabled for the runnable, then 4xx errors will be thrown
@@ -599,10 +588,6 @@ class APIHandler:
                 If not provided, then all logs will be allowed to be streamed.
                 Use to also limit the events that can be streamed by the stream_events.
                 TODO: Introduce deprecation for this parameter to rename it
-            astream_events_version: version of the stream events endpoint to use.
-                By default "v2".
-            serializer: optional serializer to use for serializing the output.
-                If not provided, the default serializer will be used.
         """
         if importlib.util.find_spec("sse_starlette") is None:
             raise ImportError(
@@ -634,18 +619,12 @@ class APIHandler:
         # and when tracing information is logged, we'll be able to see
         # traces for the path /foo/bar.
         self._run_name = self._base_url
-        if include_callback_events:
-            warn_beta(
-                message="Including callback events in the response is in beta. "
-                "This API may change in the future."
-            )
         self._include_callback_events = include_callback_events
         self._per_req_config_modifier = per_req_config_modifier
-        self._serializer = serializer or WellKnownLCSerializer()
+        self._serializer = WellKnownLCSerializer()
         self._enable_feedback_endpoint = enable_feedback_endpoint
         self._enable_public_trace_link_endpoint = enable_public_trace_link_endpoint
         self._names_in_stream_allow_list = stream_log_name_allow_list
-        self._astream_events_version = astream_events_version
 
         if token_feedback_config:
             if len(token_feedback_config["key_configs"]) != 1:
@@ -685,57 +664,15 @@ class APIHandler:
 
         model_namespace = _replace_non_alphanumeric_with_underscores(path.strip("/"))
 
-        try:
-            input_type_ = _resolve_model(
-                runnable.get_input_schema(), "Input", model_namespace
-            )
-        except Exception as e:
-            # Attempt to surface a more informative user facing error
-            raise_original_error = True
-            try:
-                if isinstance(runnable.get_input_schema(), BaseModelV1):
-                    raise_original_error = False
-                    raise ValueError(
-                        "Found an input type which is a pydantic v1 model."
-                        "Please use pydantic.BaseModel rather than "
-                        "pydantic.v1.BaseModel."
-                    )
-            finally:  # noqa
-                if raise_original_error:
-                    print(
-                        "Encountered an error while resolving the inputs of "
-                        "the Runnable. Try specifying the input type explicitly "
-                        "using the `with_types` method on the runnable.\n"
-                        "See https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.base.Runnable.html "  # noqa: E501
-                    )
-                    raise e
+        input_type_ = _resolve_model(
+            runnable.get_input_schema(), "Input", model_namespace
+        )
 
-        try:
-            output_type_ = _resolve_model(
-                runnable.get_output_schema(),
-                "Output",
-                model_namespace,
-            )
-        except Exception as e:
-            # Attempt to surface a more informative user facing error
-            raise_original_error = True
-            try:
-                if isinstance(runnable.get_output_schema(), BaseModelV1):
-                    raise_original_error = False
-                    raise ValueError(
-                        "Found an output type which is a pydantic v1 model."
-                        "Please use pydantic.BaseModel rather than "
-                        "pydantic.v1.BaseModel."
-                    )
-            finally:  # noqa
-                if raise_original_error:
-                    print(
-                        "Encountered an error while resolving the inputs of "
-                        "the Runnable. Try specifying the output type explicitly "
-                        "using the `with_types` method on the runnable.\n"
-                        "See https://api.python.langchain.com/en/latest/runnables/langchain_core.runnables.base.Runnable.html "  # noqa: E501
-                    )
-                    raise e
+        output_type_ = _resolve_model(
+            runnable.get_output_schema(),
+            "Output",
+            model_namespace,
+        )
 
         self._ConfigPayload = _add_namespace_to_model(
             model_namespace, runnable.config_schema(include=config_keys)
@@ -816,7 +753,7 @@ class APIHandler:
         except json.JSONDecodeError:
             raise RequestValidationError(errors=["Invalid JSON body"])
         try:
-            body = InvokeRequestShallowValidator.model_validate(body)
+            body = InvokeRequestShallowValidator.validate(body)
 
             # Merge the config from the path with the config from the body.
             user_provided_config = await _unpack_request_config(
@@ -838,7 +775,7 @@ class APIHandler:
             # This takes into account changes in the input type when
             # using configuration.
             schema = self._runnable.with_config(config).input_schema
-            input_ = schema.model_validate(body.input)
+            input_ = schema.validate(body.input)
             return config, _unpack_input(input_)
         except ValidationError as e:
             raise RequestValidationError(e.errors(), body=body)
@@ -948,7 +885,7 @@ class APIHandler:
             raise RequestValidationError(errors=["Invalid JSON body"])
 
         with _with_validation_error_translation():
-            body = BatchRequestShallowValidator.model_validate(body)
+            body = BatchRequestShallowValidator.validate(body)
             config = body.config
 
             # First unpack the config
@@ -999,7 +936,7 @@ class APIHandler:
 
         inputs = [
             _unpack_input(
-                self._runnable.with_config(config_).input_schema.model_validate(input_)
+                self._runnable.with_config(config_).input_schema.validate(input_)
             )
             for config_, input_ in zip(configs_, inputs_)
         ]
@@ -1399,7 +1336,7 @@ class APIHandler:
                     exclude_names=stream_events_request.exclude_names,
                     exclude_types=stream_events_request.exclude_types,
                     exclude_tags=stream_events_request.exclude_tags,
-                    version=self._astream_events_version,
+                    version="v1",
                 ):
                     if (
                         self._names_in_stream_allow_list is None
@@ -1468,7 +1405,7 @@ class APIHandler:
                 self._run_name, user_provided_config, request
             )
 
-        return self._runnable.get_input_schema(config).model_json_schema()
+        return self._runnable.get_input_schema(config).schema()
 
     async def output_schema(
         self,
@@ -1495,7 +1432,7 @@ class APIHandler:
             config = _update_config_with_defaults(
                 self._run_name, user_provided_config, request
             )
-        return self._runnable.get_output_schema(config).model_json_schema()
+        return self._runnable.get_output_schema(config).schema()
 
     async def config_schema(
         self,
@@ -1525,7 +1462,7 @@ class APIHandler:
         return (
             self._runnable.with_config(config)
             .config_schema(include=self._config_keys)
-            .model_json_schema()
+            .schema()
         )
 
     async def playground(
@@ -1653,7 +1590,6 @@ class APIHandler:
             score=create_request.score,
             value=create_request.value,
             comment=create_request.comment,
-            correction=create_request.correction,
             metadata=metadata,
         )
 

langserve/callbacks.py

@@ -48,7 +48,7 @@ class AsyncEventAggregatorCallback(AsyncCallbackHandler):
 
     async def on_chat_model_start(
         self,
-        serialized: Optional[Dict[str, Any]],
+        serialized: Dict[str, Any],
         messages: List[List[BaseMessage]],
         *,
         run_id: UUID,
@@ -73,7 +73,7 @@ class AsyncEventAggregatorCallback(AsyncCallbackHandler):
 
     async def on_chain_start(
         self,
-        serialized: Optional[Dict[str, Any]],
+        serialized: Dict[str, Any],
         inputs: Dict[str, Any],
         *,
         run_id: UUID,
@@ -138,7 +138,7 @@ class AsyncEventAggregatorCallback(AsyncCallbackHandler):
 
     async def on_retriever_start(
         self,
-        serialized: Optional[Dict[str, Any]],
+        serialized: Dict[str, Any],
         query: str,
         *,
         run_id: UUID,
@@ -202,7 +202,7 @@ class AsyncEventAggregatorCallback(AsyncCallbackHandler):
 
     async def on_tool_start(
         self,
-        serialized: Optional[Dict[str, Any]],
+        serialized: Dict[str, Any],
         input_str: str,
         *,
         run_id: UUID,
@@ -306,7 +306,7 @@ class AsyncEventAggregatorCallback(AsyncCallbackHandler):
 
     async def on_llm_start(
         self,
-        serialized: Optional[Dict[str, Any]],
+        serialized: Dict[str, Any],
         prompts: List[str],
         *,
         run_id: UUID,
@@ -445,14 +445,7 @@ async def ahandle_callbacks(
         if event["parent_run_id"] is None:  # How do we make sure it's None!?
             event["parent_run_id"] = callback_manager.run_id
 
-        event_data = {
-            key: value
-            for key, value in event.items()
-            if key != "type" and key != "kwargs"
-        }
-
-        if "kwargs" in event:
-            event_data.update(event["kwargs"])
+        event_data = {key: value for key, value in event.items() if key != "type"}
 
         await ahandle_event(
             # Unpacking like this may not work
@@ -474,14 +467,7 @@ def handle_callbacks(
         if event["parent_run_id"] is None:  # How do we make sure it's None!?
             event["parent_run_id"] = callback_manager.run_id
 
-        event_data = {
-            key: value
-            for key, value in event.items()
-            if key != "type" and key != "kwargs"
-        }
-
-        if "kwargs" in event:
-            event_data.update(event["kwargs"])
+        event_data = {key: value for key, value in event.items() if key != "type"}
 
         handle_event(
             # Unpacking like this may not work

langserve/chat_playground/dist/index.html

@@ -5,7 +5,7 @@
     <link rel="icon" href="/____LANGSERVE_BASE_URL/favicon.ico" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
     <title>Chat Playground</title>
-    <script type="module" crossorigin src="/____LANGSERVE_BASE_URL/assets/index-53ad47d4.js"></script>
+    <script type="module" crossorigin src="/____LANGSERVE_BASE_URL/assets/index-86d4d9c0.js"></script>
     <link rel="stylesheet" href="/____LANGSERVE_BASE_URL/assets/index-434ff580.css">
   </head>
   <body>

langserve/chat_playground/src/App.tsx

@@ -30,7 +30,6 @@ export function App() {
   );
   const outputSchemaSupported = (
     outputDataSchema?.anyOf?.find((option) => option.properties?.type?.enum?.includes("ai")) ||
-    outputDataSchema?.oneOf?.find((option) => option.properties?.type?.enum?.includes("ai")) ||
     outputDataSchema?.type === "string"
   );
   const isSupported = isLoading || (inputSchemaSupported && outputSchemaSupported);

langserve/client.py

@@ -120,12 +120,6 @@ def _log_error_message_once(error_message: str) -> None:
     logger.error(error_message)
 
 
-@lru_cache(maxsize=1_000)  # Will accommodate up to 1_000 different error messages
-def _log_info_message_once(error_message: str) -> None:
-    """Log an error message once."""
-    logger.info(error_message)
-
-
 def _sanitize_request(request: httpx.Request) -> httpx.Request:
     """Remove sensitive headers from the request."""
     accept_headers = {
@@ -285,7 +279,6 @@ class RemoteRunnable(Runnable[Input, Output]):
         cert: Optional[CertTypes] = None,
         client_kwargs: Optional[Dict[str, Any]] = None,
         use_server_callback_events: bool = True,
-        serializer: Optional[Serializer] = None,
     ) -> None:
         """Initialize the client.
 
@@ -301,8 +294,6 @@ class RemoteRunnable(Runnable[Input, Output]):
                 and async httpx clients
             use_server_callback_events: Whether to invoke callbacks on any
                 callback events returned by the server.
-            serializer: The serializer to use for serializing and deserializing
-                data. If not provided, a default serializer will be used.
         """
         _client_kwargs = client_kwargs or {}
         # Enforce trailing slash
@@ -330,7 +321,7 @@ class RemoteRunnable(Runnable[Input, Output]):
 
         # Register cleanup handler once RemoteRunnable is garbage collected
         weakref.finalize(self, _close_clients, self.sync_client, self.async_client)
-        self._lc_serializer = serializer or WellKnownLCSerializer()
+        self._lc_serializer = WellKnownLCSerializer()
         self._use_server_callback_events = use_server_callback_events
 
     def _invoke(
@@ -440,15 +431,11 @@ class RemoteRunnable(Runnable[Input, Output]):
         self,
         inputs: List[Input],
         config: Optional[RunnableConfig] = None,
-        *,
-        return_exceptions: bool = False,
         **kwargs: Any,
     ) -> List[Output]:
         if kwargs:
-            raise NotImplementedError(f"kwargs not implemented yet. Got {kwargs}")
-        return self._batch_with_config(
-            self._batch, inputs, config, return_exceptions=return_exceptions
-        )
+            raise NotImplementedError("kwargs not implemented yet.")
+        return self._batch_with_config(self._batch, inputs, config)
 
     async def _abatch(
         self,
@@ -761,7 +748,7 @@ class RemoteRunnable(Runnable[Input, Output]):
         input: Any,
         config: Optional[RunnableConfig] = None,
         *,
-        version: Literal["v1", "v2", None] = None,
+        version: Literal["v1"],
         include_names: Optional[Sequence[str]] = None,
         include_types: Optional[Sequence[str]] = None,
         include_tags: Optional[Sequence[str]] = None,
@@ -784,8 +771,7 @@ class RemoteRunnable(Runnable[Input, Output]):
             input: The input to the runnable
             config: The config to use for the runnable
             version: The version of the astream_events to use.
-                Currently, this input is IGNORED on the client.
-                The server will return whatever format it's configured with.
+                     Currently only "v1" is supported.
             include_names: The names of the events to include
             include_types: The types of the events to include
             include_tags: The tags of the events to include
@@ -793,18 +779,13 @@ class RemoteRunnable(Runnable[Input, Output]):
             exclude_types: The types of the events to exclude
             exclude_tags: The tags of the events to exclude
         """
+        if version != "v1":
+            raise ValueError(f"Unsupported version: {version}. Use 'v1'")
+
         # Create a stream handler that will emit Log objects
         config = ensure_config(config)
         callback_manager = get_async_callback_manager_for_config(config)
 
-        if version is not None:
-            _log_info_message_once(
-                "Versioning of the astream_events API is not supported on the client "
-                "side currently. The server will return events in whatever format "
-                "it was configured with in add_routes or APIHandler. "
-                "To stop seeing this message, remove the `version` argument."
-            )
-
         events = []
 
         run_manager = await callback_manager.on_chain_start(

langserve/playground.py

@@ -6,7 +6,8 @@ from typing import Literal, Sequence, Type
 
 from fastapi.responses import Response
 from langchain_core.runnables import Runnable
-from pydantic import BaseModel
+
+from langserve.pydantic_v1 import BaseModel
 
 
 class PlaygroundTemplate(Template):
@@ -89,12 +90,10 @@ async def serve_playground(
                     if base_url.startswith("/")
                     else base_url,
                     LANGSERVE_CONFIG_SCHEMA=json.dumps(
-                        runnable.config_schema(include=config_keys).model_json_schema()
-                    ),
-                    LANGSERVE_INPUT_SCHEMA=json.dumps(input_schema.model_json_schema()),
-                    LANGSERVE_OUTPUT_SCHEMA=json.dumps(
-                        output_schema.model_json_schema()
+                        runnable.config_schema(include=config_keys).schema()
                     ),
+                    LANGSERVE_INPUT_SCHEMA=json.dumps(input_schema.schema()),
+                    LANGSERVE_OUTPUT_SCHEMA=json.dumps(output_schema.schema()),
                     LANGSERVE_FEEDBACK_ENABLED=json.dumps(
                         "true" if feedback_enabled else "false"
                     ),

langserve/playground/dist/index.html

@@ -5,7 +5,7 @@
     <link rel="icon" href="/____LANGSERVE_BASE_URL/favicon.ico" />
     <meta name="viewport" content="width=device-width, initial-scale=1.0" />
     <title>Playground</title>
-    <script type="module" crossorigin src="/____LANGSERVE_BASE_URL/assets/index-400979f0.js"></script>
+    <script type="module" crossorigin src="/____LANGSERVE_BASE_URL/assets/index-dbc96538.js"></script>
     <link rel="stylesheet" href="/____LANGSERVE_BASE_URL/assets/index-52e8ab2f.css">
   </head>
   <body>

langserve/playground/src/components/ChatMessagesControlRenderer.tsx

@@ -6,30 +6,12 @@ import {
   schemaMatches,
   Paths,
   isControl,
-  JsonSchema,
 } from "@jsonforms/core";
 import { useStreamCallback } from "../useStreamCallback";
 import { isJsonSchemaExtra } from "../utils/schema";
 import { MessageFields, ChatMessageInput } from "./ChatMessageInput";
 import { useEffect } from "react";
 
-function checkItemSchema(schema: JsonSchema) {
-  const isObjectMessage =
-    schema.type === "object" &&
-    (schema.title?.endsWith("Message") ||
-      schema.title?.endsWith("MessageChunk"));
-
-  const isTupleMessage =
-    schema.type === "array" &&
-    schema.minItems === 2 &&
-    schema.maxItems === 2 &&
-    Array.isArray(schema.items) &&
-    schema.items.length === 2 &&
-    schema.items.every((schema) => schema.type === "string");
-
-  return isObjectMessage || isTupleMessage;
-}
-
 export const chatMessagesTester = rankWith(
   12,
   and(
@@ -52,11 +34,22 @@ export const chatMessagesTester = rankWith(
       }
 
       if ("anyOf" in schema.items && schema.items.anyOf != null) {
-        return schema.items.anyOf.every(checkItemSchema);
-      }
+        return schema.items.anyOf.every((schema) => {
+          const isObjectMessage =
+            schema.type === "object" &&
+            (schema.title?.endsWith("Message") ||
+              schema.title?.endsWith("MessageChunk"));
 
-      if ("oneOf" in schema.items && schema.items.oneOf != null) {
-        return schema.items.oneOf.every(checkItemSchema);
+          const isTupleMessage =
+            schema.type === "array" &&
+            schema.minItems === 2 &&
+            schema.maxItems === 2 &&
+            Array.isArray(schema.items) &&
+            schema.items.length === 2 &&
+            schema.items.every((schema) => schema.type === "string");
+
+          return isObjectMessage || isTupleMessage;
+        });
       }
 
       return false;
@@ -71,14 +64,10 @@ export const ChatMessagesControlRenderer = withJsonFormsControlProps(
     useEffect(() => {
       if (!isJsonSchemaExtra(props.schema)) return;
       if (props.schema.extra.widget.type !== "chat") return;
-      setTimeout(
-        () =>
-          props.handleChange(props.path, [
-            ...data,
-            { content: "", type: "human" },
-          ]),
-        10
-      );
+      setTimeout(() => props.handleChange(props.path, [
+        ...data,
+        { content: "", type: "human" },
+      ]), 10);
     }, []);
 
     useStreamCallback("onStart", () => {
@@ -92,10 +81,7 @@ export const ChatMessagesControlRenderer = withJsonFormsControlProps(
       if (props.schema.extra.widget.type !== "chat") return;
       if (aggregatedState?.final_output !== undefined) {
         const msgPath = Paths.compose(props.path, `${data.length - 1}`);
-        if (
-          (aggregatedState.final_output as MessageFields)?.type ===
-          "AIMessageChunk"
-        ) {
+        if ((aggregatedState.final_output as MessageFields)?.type === "AIMessageChunk") {
           props.handleChange(
             Paths.compose(msgPath, "content"),
             (aggregatedState.final_output as MessageFields)?.content
@@ -154,7 +140,7 @@ export const ChatMessagesControlRenderer = withJsonFormsControlProps(
                 props.path,
                 data.filter((_, i) => i !== index)
               );
-            };
+            }
             return (
               <ChatMessageInput
                 message={message}
@@ -162,7 +148,7 @@ export const ChatMessagesControlRenderer = withJsonFormsControlProps(
                 handleRemoval={handleChatMessageRemoval}
                 path={props.path}
                 key={index}
-              ></ChatMessageInput>
+                ></ChatMessageInput>
             );
           })}
         </div>

langserve/pydantic_v1.py

@@ -0,0 +1,33 @@
+from importlib import metadata
+
+## Create namespaces for pydantic v1 and v2.
+# This code must stay at the top of the file before other modules may
+# attempt to import pydantic since it adds pydantic_v1 and pydantic_v2 to sys.modules.
+#
+# This hack is done for the following reasons:
+# * Langchain will attempt to remain compatible with both pydantic v1 and v2 since
+#   both dependencies and dependents may be stuck on either version of v1 or v2.
+# * Creating namespaces for pydantic v1 and v2 should allow us to write code that
+#   unambiguously uses either v1 or v2 API.
+# * This change is easier to roll out and roll back.
+
+try:
+    # F401: imported but unused
+    from pydantic.v1 import (  # noqa: F401
+        BaseModel,
+        Field,
+        ValidationError,
+        create_model,
+    )
+except ImportError:
+    from pydantic import BaseModel, Field, ValidationError, create_model  # noqa: F401
+
+
+# This is not a pydantic v1 thing, but it feels too small to create a new module for.
+
+PYDANTIC_VERSION = metadata.version("pydantic")
+
+try:
+    _PYDANTIC_MAJOR_VERSION: int = int(PYDANTIC_VERSION.split(".")[0])
+except metadata.PackageNotFoundError:
+    _PYDANTIC_MAJOR_VERSION = -1

langserve/schema.py

@@ -2,11 +2,10 @@ from datetime import datetime
 from typing import Dict, List, Optional, Union
 from uuid import UUID
 
-from pydantic import (
-    BaseModel,
-    Field,
-)
-from pydantic import BaseModel as BaseModelV1
+from pydantic import BaseModel  # Floats between v1 and v2
+
+from langserve.pydantic_v1 import BaseModel as BaseModelV1
+from langserve.pydantic_v1 import Field
 
 
 class CustomUserType(BaseModelV1):

langserve/serialization.py

@@ -13,7 +13,7 @@ sensitive information from the server to the client.
 import abc
 import logging
 from functools import lru_cache
-from typing import Annotated, Any, Dict, List, Union
+from typing import Any, Dict, List, Union
 
 import orjson
 from langchain_core.agents import AgentAction, AgentActionMessageLog, AgentFinish
@@ -29,8 +29,6 @@ from langchain_core.messages import (
     HumanMessageChunk,
     SystemMessage,
     SystemMessageChunk,
-    ToolMessage,
-    ToolMessageChunk,
 )
 from langchain_core.outputs import (
     ChatGeneration,
@@ -40,8 +38,8 @@ from langchain_core.outputs import (
 )
 from langchain_core.prompt_values import ChatPromptValueConcrete
 from langchain_core.prompts.base import StringPromptValue
-from pydantic import BaseModel, Discriminator, Field, RootModel, Tag, ValidationError
 
+from langserve.pydantic_v1 import BaseModel, ValidationError
 from langserve.validation import CallbackEvent
 
 logger = logging.getLogger(__name__)
@@ -53,58 +51,42 @@ def _log_error_message_once(error_message: str) -> None:
     logger.error(error_message)
 
 
-def _get_type(v: Any) -> str:
-    """Get the type associated with the object for serialization purposes."""
-    if isinstance(v, dict) and "type" in v:
-        return v["type"]
-    elif hasattr(v, "type"):
-        return v.type
-    else:
-        raise TypeError(
-            f"Expected either a dictionary with a 'type' key or an object "
-            f"with a 'type' attribute. Instead got type {type(v)}."
-        )
+class WellKnownLCObject(BaseModel):
+    """A well known LangChain object.
 
+    A pydantic model that defines what constitutes a well known LangChain object.
 
-# A well known LangChain object.
-# A pydantic model that defines what constitutes a well known LangChain object.
-# All well-known objects are allowed to be serialized and de-serialized.
+    All well-known objects are allowed to be serialized and de-serialized.
+    """
 
-WellKnownLCObject = RootModel[
-    Annotated[
-        Union[
-            Annotated[AIMessage, Tag(tag="ai")],
-            Annotated[HumanMessage, Tag(tag="human")],
-            Annotated[ChatMessage, Tag(tag="chat")],
-            Annotated[SystemMessage, Tag(tag="system")],
-            Annotated[FunctionMessage, Tag(tag="function")],
-            Annotated[ToolMessage, Tag(tag="tool")],
-            Annotated[AIMessageChunk, Tag(tag="AIMessageChunk")],
-            Annotated[HumanMessageChunk, Tag(tag="HumanMessageChunk")],
-            Annotated[ChatMessageChunk, Tag(tag="ChatMessageChunk")],
-            Annotated[SystemMessageChunk, Tag(tag="SystemMessageChunk")],
-            Annotated[FunctionMessageChunk, Tag(tag="FunctionMessageChunk")],
-            Annotated[ToolMessageChunk, Tag(tag="ToolMessageChunk")],
-            Annotated[Document, Tag(tag="Document")],
-            Annotated[StringPromptValue, Tag(tag="StringPromptValue")],
-            Annotated[ChatPromptValueConcrete, Tag(tag="ChatPromptValueConcrete")],
-            Annotated[AgentAction, Tag(tag="AgentAction")],
-            Annotated[AgentFinish, Tag(tag="AgentFinish")],
-            Annotated[AgentActionMessageLog, Tag(tag="AgentActionMessageLog")],
-            Annotated[ChatGeneration, Tag(tag="ChatGeneration")],
-            Annotated[Generation, Tag(tag="Generation")],
-            Annotated[ChatGenerationChunk, Tag(tag="ChatGenerationChunk")],
-            Annotated[LLMResult, Tag(tag="LLMResult")],
-        ],
-        Field(discriminator=Discriminator(_get_type)),
+    __root__: Union[
+        Document,
+        HumanMessage,
+        SystemMessage,
+        ChatMessage,
+        FunctionMessage,
+        AIMessage,
+        HumanMessageChunk,
+        SystemMessageChunk,
+        ChatMessageChunk,
+        FunctionMessageChunk,
+        AIMessageChunk,
+        StringPromptValue,
+        ChatPromptValueConcrete,
+        AgentAction,
+        AgentFinish,
+        AgentActionMessageLog,
+        LLMResult,
+        ChatGeneration,
+        Generation,
+        ChatGenerationChunk,
     ]
-]
 
 
 def default(obj) -> Any:
     """Default serialization for well known objects."""
     if isinstance(obj, BaseModel):
-        return obj.model_dump()
+        return obj.dict()
     return super().default(obj)
 
 
@@ -114,10 +96,12 @@ def _decode_lc_objects(value: Any) -> Any:
         v = {key: _decode_lc_objects(v) for key, v in value.items()}
 
         try:
-            obj = WellKnownLCObject.model_validate(v)
-            parsed = obj.root
+            obj = WellKnownLCObject.parse_obj(v)
+            parsed = obj.__root__
+            if set(parsed.dict()) != set(value):
+                raise ValueError("Invalid object")
             return parsed
-        except (ValidationError, ValueError, TypeError):
+        except (ValidationError, ValueError):
             return v
     elif isinstance(value, list):
         return [_decode_lc_objects(item) for item in value]
@@ -139,12 +123,12 @@ def _decode_event_data(value: Any) -> Any:
     """Decode the event data from a JSON object representation."""
     if isinstance(value, dict):
         try:
-            obj = CallbackEvent.model_validate(value)
-            return obj.root
+            obj = CallbackEvent.parse_obj(value)
+            return obj.__root__
         except ValidationError:
             try:
-                obj = WellKnownLCObject.model_validate(value)
-                return obj.root
+                obj = WellKnownLCObject.parse_obj(value)
+                return obj.__root__
             except ValidationError:
                 return {key: _decode_event_data(v) for key, v in value.items()}
     elif isinstance(value, list):
@@ -157,54 +141,44 @@ def _decode_event_data(value: Any) -> Any:
 
 
 class Serializer(abc.ABC):
+    @abc.abstractmethod
     def dumpd(self, obj: Any) -> Any:
         """Convert the given object to a JSON serializable object."""
-        return orjson.loads(self.dumps(obj))
+
+    @abc.abstractmethod
+    def dumps(self, obj: Any) -> bytes:
+        """Dump the given object as a JSON string."""
+
+    @abc.abstractmethod
+    def loads(self, s: bytes) -> Any:
+        """Load the given JSON string."""
+
+    @abc.abstractmethod
+    def loadd(self, obj: Any) -> Any:
+        """Load the given object."""
+
+
+class WellKnownLCSerializer(Serializer):
+    def dumpd(self, obj: Any) -> Any:
+        """Convert the given object to a JSON serializable object."""
+        return orjson.loads(orjson.dumps(obj, default=default))
+
+    def dumps(self, obj: Any) -> bytes:
+        """Dump the given object as a JSON string."""
+        return orjson.dumps(obj, default=default)
+
+    def loadd(self, obj: Any) -> Any:
+        """Load the given object."""
+        return _decode_lc_objects(obj)
 
     def loads(self, s: bytes) -> Any:
         """Load the given JSON string."""
         return self.loadd(orjson.loads(s))
 
-    @abc.abstractmethod
-    def dumps(self, obj: Any) -> bytes:
-        """Dump the given object to a JSON byte string."""
-
-    @abc.abstractmethod
-    def loadd(self, s: bytes) -> Any:
-        """Given a python object, load it into a well known object.
-
-        The obj represents content that was json loaded from a string, but
-        not yet validated or converted into a well known object.
-        """
-
-
-class WellKnownLCSerializer(Serializer):
-    """A pre-defined serializer for well known LangChain objects.
-
-    This is the default serialized used by LangServe for serializing and
-    de-serializing well known LangChain objects.
-
-    If you need to extend the serialization capabilities for your own application,
-    feel free to create a new instance of the Serializer class and implement
-    the abstract methods dumps and loadd.
-    """
-
-    def dumps(self, obj: Any) -> bytes:
-        """Dump the given object to a JSON byte string."""
-        return orjson.dumps(obj, default=default)
-
-    def loadd(self, obj: Any) -> Any:
-        """Given a python object, load it into a well known object.
-
-        The obj represents content that was json loaded from a string, but
-        not yet validated or converted into a well known object.
-        """
-        return _decode_lc_objects(obj)
-
 
 def _project_top_level(model: BaseModel) -> Dict[str, Any]:
     """Project the top level of the model as dict."""
-    return {key: getattr(model, key) for key in model.model_fields}
+    return {key: getattr(model, key) for key in model.__fields__}
 
 
 def load_events(events: Any) -> List[Dict[str, Any]]:
@@ -235,7 +209,7 @@ def load_events(events: Any) -> List[Dict[str, Any]]:
 
         # Then validate the event
         try:
-            full_event = CallbackEvent.model_validate(decoded_event_data)
+            full_event = CallbackEvent.parse_obj(decoded_event_data)
         except ValidationError as e:
             msg = f"Encountered an invalid event: {e}"
             if "type" in decoded_event_data:
@@ -243,7 +217,7 @@ def load_events(events: Any) -> List[Dict[str, Any]]:
             _log_error_message_once(msg)
             continue
 
-        decoded_event_data = _project_top_level(full_event.root)
+        decoded_event_data = _project_top_level(full_event.__root__)
 
         if decoded_event_data["type"].endswith("_error"):
             # Data is validated by this point, so we can assume that the shape

langserve/server.py

@@ -5,7 +5,6 @@ This code contains integration for langchain runnables with FastAPI.
 The main entry point is the `add_routes` function which adds the routes to an existing
 FastAPI app or APIRouter.
 """
-import warnings
 import weakref
 from typing import (
     Any,
@@ -17,7 +16,6 @@ from typing import (
 )
 
 from langchain_core.runnables import Runnable
-from pydantic import BaseModel
 from typing_extensions import Annotated
 
 from langserve.api_handler import (
@@ -26,7 +24,11 @@ from langserve.api_handler import (
     TokenFeedbackConfig,
     _is_hosted,
 )
-from langserve.serialization import Serializer
+from langserve.pydantic_v1 import (
+    _PYDANTIC_MAJOR_VERSION,
+    PYDANTIC_VERSION,
+    BaseModel,
+)
 
 try:
     from fastapi import APIRouter, Depends, FastAPI, Request, Response
@@ -203,43 +205,51 @@ def _register_path_for_app(
 def _setup_global_app_handlers(
     app: Union[FastAPI, APIRouter], endpoint_configuration: _EndpointConfiguration
 ) -> None:
-    with warnings.catch_warnings():
-        # We are using deprecated functionality here.
-        # We should re-write to use lifetime events at some point, and yielding
-        # an APIRouter instance to the caller.
-        warnings.filterwarnings(
-            "ignore",
-            "[\\s.]*on_event is deprecated[\\s.]*",
-            category=DeprecationWarning,
-        )
+    @app.on_event("startup")
+    async def startup_event():
+        LANGSERVE = r"""
+ __          ___      .__   __.   _______      _______. _______ .______     ____    ____  _______
+|  |        /   \     |  \ |  |  /  _____|    /       ||   ____||   _  \    \   \  /   / |   ____|
+|  |       /  ^  \    |   \|  | |  |  __     |   (----`|  |__   |  |_)  |    \   \/   /  |  |__
+|  |      /  /_\  \   |  . `  | |  | |_ |     \   \    |   __|  |      /      \      /   |   __|
+|  `----./  _____  \  |  |\   | |  |__| | .----)   |   |  |____ |  |\  \----.  \    /    |  |____
+|_______/__/     \__\ |__| \__|  \______| |_______/    |_______|| _| `._____|   \__/     |_______|
+"""  # noqa: E501
 
-        @app.on_event("startup")
-        async def startup_event():
-            LANGSERVE = r"""
-     __          ___      .__   __.   _______      _______. _______ .______     ____    ____  _______
-    |  |        /   \     |  \ |  |  /  _____|    /       ||   ____||   _  \    \   \  /   / |   ____|
-    |  |       /  ^  \    |   \|  | |  |  __     |   (----`|  |__   |  |_)  |    \   \/   /  |  |__
-    |  |      /  /_\  \   |  . `  | |  | |_ |     \   \    |   __|  |      /      \      /   |   __|
-    |  `----./  _____  \  |  |\   | |  |__| | .----)   |   |  |____ |  |\  \----.  \    /    |  |____
-    |_______/__/     \__\ |__| \__|  \______| |_______/    |_______|| _| `._____|   \__/     |_______|
-    """  # noqa: E501
+        def green(text: str) -> str:
+            """Return the given text in green."""
+            return "\x1b[1;32;40m" + text + "\x1b[0m"
 
-            def green(text: str) -> str:
-                """Return the given text in green."""
-                return "\x1b[1;32;40m" + text + "\x1b[0m"
+        def orange(text: str) -> str:
+            """Return the given text in orange."""
+            return "\x1b[1;31;40m" + text + "\x1b[0m"
 
-            paths = _APP_TO_PATHS[app]
-            print(LANGSERVE)
-            for path in paths:
-                if endpoint_configuration.is_playground_enabled:
-                    print(
-                        f'{green("LANGSERVE:")} Playground for chain "{path or ""}/" '
-                        f'is live at:'
-                    )
-                    print(f'{green("LANGSERVE:")}  │')
-                    print(f'{green("LANGSERVE:")}  └──> {path}/playground/')
-                    print(f'{green("LANGSERVE:")}')
-            print(f'{green("LANGSERVE:")} See all available routes at {app.docs_url}/')
+        paths = _APP_TO_PATHS[app]
+        print(LANGSERVE)
+        for path in paths:
+            if endpoint_configuration.is_playground_enabled:
+                print(
+                    f'{green("LANGSERVE:")} Playground for chain "{path or ""}/" is '
+                    f"live at:"
+                )
+                print(f'{green("LANGSERVE:")}  │')
+                print(f'{green("LANGSERVE:")}  └──> {path}/playground/')
+                print(f'{green("LANGSERVE:")}')
+        print(f'{green("LANGSERVE:")} See all available routes at {app.docs_url}/')
+
+        if _PYDANTIC_MAJOR_VERSION == 2:
+            print()
+            print(f'{orange("LANGSERVE:")} ', end="")
+            print(
+                f"⚠️ Using pydantic {PYDANTIC_VERSION}. "
+                f"OpenAPI docs for invoke, batch, stream, stream_log "
+                f"endpoints will not be generated. API endpoints and playground "
+                f"should work as expected. "
+                f"If you need to see the docs, you can downgrade to pydantic 1. "
+                "For example, `pip install pydantic==1.10.13`. "
+                f"See https://github.com/tiangolo/fastapi/issues/10360 for details."
+            )
+        print()
 
 
 # PUBLIC API
@@ -263,8 +273,6 @@ def add_routes(
     enabled_endpoints: Optional[Sequence[EndpointName]] = None,
     dependencies: Optional[Sequence[Depends]] = None,
     playground_type: Literal["default", "chat"] = "default",
-    astream_events_version: Literal["v1", "v2"] = "v2",
-    serializer: Optional[Serializer] = None,
 ) -> None:
     """Register the routes on the given FastAPI app or APIRouter.
 
@@ -383,18 +391,14 @@ def add_routes(
             - chat: UX is optimized for chat-like interactions. Please review
               the README in langserve for more details about constraints (e.g.,
               which message types are supported etc.)
-        astream_events_version: version of the stream events endpoint to use.
-            By default "v2".
-        serializer: The serializer to use for serializing the output. If not provided,
-            the default serializer will be used.
     """  # noqa: E501
     if not isinstance(runnable, Runnable):
         raise TypeError(
             f"Expected a Runnable, got {type(runnable)}. "
-            "The second argument to add_routes should be a Runnable instance."
-            "add_route(app, runnable, ...) is the correct usage."
-            "Please make sure that you are using a runnable which is an instance of "
-            "langchain_core.runnables.Runnable."
+            f"The second argument to add_routes should be a Runnable instance."
+            f"add_route(app, runnable, ...) is the correct usage."
+            f"Please make sure that you are using a runnable which is an instance of "
+            f"langchain_core.runnables.Runnable."
         )
 
     endpoint_configuration = _EndpointConfiguration(
@@ -450,10 +454,7 @@ def add_routes(
         per_req_config_modifier=per_req_config_modifier,
         stream_log_name_allow_list=stream_log_name_allow_list,
         playground_type=playground_type,
-        astream_events_version=astream_events_version,
-        serializer=serializer,
     )
-
     namespace = path or ""
 
     route_tags = [path.strip("/")] if path else None
@@ -472,9 +473,35 @@ def add_routes(
     if hasattr(app, "openapi_tags") and (path or (app not in _APP_SEEN)):
         if not path:
             _APP_SEEN.add(app)
-        default_endpoint_tags = {
-            "name": route_tags[0] if route_tags else "default",
-        }
+
+        if _PYDANTIC_MAJOR_VERSION == 1:
+            # Documentation for the default endpoints
+            default_endpoint_tags = {
+                "name": route_tags[0] if route_tags else "default",
+            }
+        elif _PYDANTIC_MAJOR_VERSION == 2:
+            # When using pydantic v2, we cannot generate openapi docs for
+            # the invoke/batch/stream/stream_log endpoints since the underlying
+            # models are from the pydantic.v1 namespace and cannot be supported
+            # by FastAPI's.
+            # https://github.com/tiangolo/fastapi/issues/10360
+            default_endpoint_tags = {
+                "name": route_tags[0] if route_tags else "default",
+                "description": (
+                    f"⚠️ Using pydantic {PYDANTIC_VERSION}. "
+                    f"OpenAPI docs for `invoke`, `batch`, `stream`, `stream_log` "
+                    f"endpoints will not be generated. API endpoints and playground "
+                    f"should work as expected. "
+                    f"If you need to see the docs, you can downgrade to pydantic 1. "
+                    "For example, `pip install pydantic==1.10.13`"
+                    f"See https://github.com/tiangolo/fastapi/issues/10360 for details."
+                ),
+            }
+        else:
+            raise AssertionError(
+                f"Expected pydantic major version 1 or 2, got {_PYDANTIC_MAJOR_VERSION}"
+            )
+
         if endpoint_configuration.is_config_hash_enabled:
             app.openapi_tags = [
                 *(getattr(app, "openapi_tags", []) or []),
@@ -751,326 +778,329 @@ def add_routes(
     # Documentation variants of end points.
     #######################################
     # At the moment, we only support pydantic 1.x for documentation
-    InvokeRequest = api_handler.InvokeRequest
-    InvokeResponse = api_handler.InvokeResponse
-    BatchRequest = api_handler.BatchRequest
-    BatchResponse = api_handler.BatchResponse
-    StreamRequest = api_handler.StreamRequest
-    StreamLogRequest = api_handler.StreamLogRequest
-    StreamEventsRequest = api_handler.StreamEventsRequest
+    if _PYDANTIC_MAJOR_VERSION == 1:
+        InvokeRequest = api_handler.InvokeRequest
+        InvokeResponse = api_handler.InvokeResponse
+        BatchRequest = api_handler.BatchRequest
+        BatchResponse = api_handler.BatchResponse
+        StreamRequest = api_handler.StreamRequest
+        StreamLogRequest = api_handler.StreamLogRequest
+        StreamEventsRequest = api_handler.StreamEventsRequest
 
-    if endpoint_configuration.is_invoke_enabled:
+        if endpoint_configuration.is_invoke_enabled:
 
-        async def _invoke_docs(
-            invoke_request: Annotated[InvokeRequest, InvokeRequest],
-            config_hash: str = "",
-        ) -> InvokeResponse:
-            """Invoke the runnable with the given input and config."""
-            raise AssertionError("This endpoint should not be reachable.")
+            async def _invoke_docs(
+                invoke_request: Annotated[InvokeRequest, InvokeRequest],
+                config_hash: str = "",
+            ) -> InvokeResponse:
+                """Invoke the runnable with the given input and config."""
+                raise AssertionError("This endpoint should not be reachable.")
 
-        invoke_docs = app.post(
-            f"{namespace}/invoke",
-            response_model=api_handler.InvokeResponse,
-            tags=route_tags,
-            name=_route_name("invoke"),
-            dependencies=dependencies,
-        )(_invoke_docs)
-
-        if endpoint_configuration.is_config_hash_enabled:
-            app.post(
-                namespace + "/c/{config_hash}/invoke",
+            invoke_docs = app.post(
+                f"{namespace}/invoke",
                 response_model=api_handler.InvokeResponse,
-                tags=route_tags_with_config,
-                name=_route_name_with_config("invoke"),
+                tags=route_tags,
+                name=_route_name("invoke"),
                 dependencies=dependencies,
-                description=(
-                    "This endpoint is to be used with share links generated by the "
-                    "LangServe playground. "
-                    "The hash is an LZString compressed JSON string. "
-                    "For regular use cases, use the /invoke endpoint without "
-                    "the `c/{config_hash}` path parameter."
-                ),
-            )(invoke_docs)
+            )(_invoke_docs)
 
-    if endpoint_configuration.is_batch_enabled:
+            if endpoint_configuration.is_config_hash_enabled:
+                app.post(
+                    namespace + "/c/{config_hash}/invoke",
+                    response_model=api_handler.InvokeResponse,
+                    tags=route_tags_with_config,
+                    name=_route_name_with_config("invoke"),
+                    dependencies=dependencies,
+                    description=(
+                        "This endpoint is to be used with share links generated by the "
+                        "LangServe playground. "
+                        "The hash is an LZString compressed JSON string. "
+                        "For regular use cases, use the /invoke endpoint without "
+                        "the `c/{config_hash}` path parameter."
+                    ),
+                )(invoke_docs)
 
-        async def _batch_docs(
-            batch_request: Annotated[BatchRequest, BatchRequest],
-            config_hash: str = "",
-        ) -> BatchResponse:
-            """Batch invoke the runnable with the given inputs and config."""
-            raise AssertionError("This endpoint should not be reachable.")
+        if endpoint_configuration.is_batch_enabled:
 
-        batch_docs = app.post(
-            f"{namespace}/batch",
-            response_model=BatchResponse,
-            tags=route_tags,
-            name=_route_name("batch"),
-            dependencies=dependencies,
-        )(_batch_docs)
+            async def _batch_docs(
+                batch_request: Annotated[BatchRequest, BatchRequest],
+                config_hash: str = "",
+            ) -> BatchResponse:
+                """Batch invoke the runnable with the given inputs and config."""
+                raise AssertionError("This endpoint should not be reachable.")
 
-        if endpoint_configuration.is_config_hash_enabled:
-            app.post(
-                namespace + "/c/{config_hash}/batch",
+            batch_docs = app.post(
+                f"{namespace}/batch",
                 response_model=BatchResponse,
-                tags=route_tags_with_config,
-                name=_route_name_with_config("batch"),
+                tags=route_tags,
+                name=_route_name("batch"),
                 dependencies=dependencies,
-                description=(
-                    "This endpoint is to be used with share links generated by the "
-                    "LangServe playground. "
-                    "The hash is an LZString compressed JSON string. "
-                    "For regular use cases, use the /batch endpoint without "
-                    "the `c/{config_hash}` path parameter."
-                ),
-            )(batch_docs)
+            )(_batch_docs)
 
-    if endpoint_configuration.is_stream_enabled:
+            if endpoint_configuration.is_config_hash_enabled:
+                app.post(
+                    namespace + "/c/{config_hash}/batch",
+                    response_model=BatchResponse,
+                    tags=route_tags_with_config,
+                    name=_route_name_with_config("batch"),
+                    dependencies=dependencies,
+                    description=(
+                        "This endpoint is to be used with share links generated by the "
+                        "LangServe playground. "
+                        "The hash is an LZString compressed JSON string. "
+                        "For regular use cases, use the /batch endpoint without "
+                        "the `c/{config_hash}` path parameter."
+                    ),
+                )(batch_docs)
 
-        async def _stream_docs(
-            stream_request: Annotated[StreamRequest, StreamRequest],
-            config_hash: str = "",
-        ) -> EventSourceResponse:
-            """Invoke the runnable stream the output.
+        if endpoint_configuration.is_stream_enabled:
 
-            This endpoint allows to stream the output of the runnable.
+            async def _stream_docs(
+                stream_request: Annotated[StreamRequest, StreamRequest],
+                config_hash: str = "",
+            ) -> EventSourceResponse:
+                """Invoke the runnable stream the output.
 
-            The endpoint uses a server sent event stream to stream the output.
+                This endpoint allows to stream the output of the runnable.
 
-            https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
+                The endpoint uses a server sent event stream to stream the output.
 
-            Important: Set the "text/event-stream" media type for request headers if
-                not using an existing SDK.
+                https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
 
-            The events that the endpoint uses are the following:
-            * "data" -- used for streaming the output of the runnale
-            * "error" -- signaling an error while streaming and ends the stream.
-            * "end" -- used for signaling the end of the stream
-            * "metadata" -- used for sending metadata about the run; e.g., run id.
+                Important: Set the "text/event-stream" media type for request headers if
+                    not using an existing SDK.
 
-            The event type is in the "event" field of the event.
-            The payload associated with the event is in the "data" field
-            of the event, and it is JSON encoded.
+                The events that the endpoint uses are the following:
+                * "data" -- used for streaming the output of the runnale
+                * "error" -- signaling an error while streaming and ends the stream.
+                * "end" -- used for signaling the end of the stream
+                * "metadata" -- used for sending metadata about the run; e.g., run id.
+
+                The event type is in the "event" field of the event.
+                The payload associated with the event is in the "data" field
+                of the event, and it is JSON encoded.
 
 
-            Here are some examples of events that the endpoint can send:
-
-            Regular streaming event:
-            {
-                "event": "data",
-                "data": {
-                    ...
-                }
-            }
-
-            Internal server error:
-            {
-                "event": "error",
-                "data": {
-                    "status_code": 500,
-                    "message": "Internal Server Error"
-                }
-            }
-
-            Streaming ended so client should stop listening for events:
-            {
-                "event": "end",
-            }
-            """
-            raise AssertionError("This endpoint should not be reachable.")
-
-        stream_docs = app.post(
-            f"{namespace}/stream",
-            include_in_schema=True,
-            tags=route_tags,
-            name=_route_name("stream"),
-            dependencies=dependencies,
-            description=(
-                "This endpoint allows to stream the output of the runnable. "
-                "The endpoint uses a server sent event stream to stream the "
-                "output. "
-                "https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events"
-            ),
-        )(_stream_docs)
-
-        if endpoint_configuration.is_config_hash_enabled:
-            app.post(
-                namespace + "/c/{config_hash}/stream",
-                include_in_schema=True,
-                tags=route_tags_with_config,
-                name=_route_name_with_config("stream"),
-                dependencies=dependencies,
-                description=(
-                    "This endpoint is to be used with share links generated by the "
-                    "LangServe playground. "
-                    "The hash is an LZString compressed JSON string. "
-                    "For regular use cases, use the /stream endpoint without "
-                    "the `c/{config_hash}` path parameter."
-                ),
-            )(stream_docs)
-
-    if endpoint_configuration.is_stream_log_enabled:
-
-        async def _stream_log_docs(
-            stream_log_request: Annotated[StreamLogRequest, StreamLogRequest],
-            config_hash: str = "",
-        ) -> EventSourceResponse:
-            """Invoke the runnable stream_log the output.
-
-            This endpoint allows to stream the output of the runnable, including
-            the output of all intermediate steps.
-
-            The endpoint uses a server sent event stream to stream the output.
-
-            https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
-
-            Important: Set the "text/event-stream" media type for request headers if
-                       not using an existing SDK.
-
-            This endpoint uses two different types of events:
-
-            * data - for streaming the output of the runnable
+                Here are some examples of events that the endpoint can send:
 
+                Regular streaming event:
                 {
                     "event": "data",
                     "data": {
-                    ...
+                        ...
                     }
                 }
 
-            * error - for signaling an error in the stream, also ends the stream.
-
-            {
-                "event": "error",
-                "data": {
-                    "status_code": 500,
-                    "message": "Internal Server Error"
+                Internal server error:
+                {
+                    "event": "error",
+                    "data": {
+                        "status_code": 500,
+                        "message": "Internal Server Error"
+                    }
                 }
-            }
-
-            * end - for signaling the end of the stream.
-
-                This helps the client to know when to stop listening for events and
-                know that the streaming has ended successfully.
 
+                Streaming ended so client should stop listening for events:
                 {
                     "event": "end",
                 }
-            """
-            raise AssertionError("This endpoint should not be reachable.")
+                """
+                raise AssertionError("This endpoint should not be reachable.")
 
-        app.post(
-            f"{namespace}/stream_log",
-            include_in_schema=True,
-            tags=route_tags,
-            name=_route_name("stream_log"),
-            dependencies=dependencies,
-        )(_stream_log_docs)
-
-        if endpoint_configuration.is_config_hash_enabled:
-            app.post(
-                namespace + "/c/{config_hash}/stream_log",
+            stream_docs = app.post(
+                f"{namespace}/stream",
                 include_in_schema=True,
-                tags=route_tags_with_config,
-                name=_route_name_with_config("stream_log"),
+                tags=route_tags,
+                name=_route_name("stream"),
+                dependencies=dependencies,
                 description=(
-                    "This endpoint is to be used with share links generated by the "
-                    "LangServe playground. "
-                    "The hash is an LZString compressed JSON string. "
-                    "For regular use cases, use the /stream_log endpoint without "
-                    "the `c/{config_hash}` path parameter."
+                    "This endpoint allows to stream the output of the runnable. "
+                    "The endpoint uses a server sent event stream to stream the "
+                    "output. "
+                    "https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events"
                 ),
+            )(_stream_docs)
+
+            if endpoint_configuration.is_config_hash_enabled:
+                app.post(
+                    namespace + "/c/{config_hash}/stream",
+                    include_in_schema=True,
+                    tags=route_tags_with_config,
+                    name=_route_name_with_config("stream"),
+                    dependencies=dependencies,
+                    description=(
+                        "This endpoint is to be used with share links generated by the "
+                        "LangServe playground. "
+                        "The hash is an LZString compressed JSON string. "
+                        "For regular use cases, use the /stream endpoint without "
+                        "the `c/{config_hash}` path parameter."
+                    ),
+                )(stream_docs)
+
+        if endpoint_configuration.is_stream_log_enabled:
+
+            async def _stream_log_docs(
+                stream_log_request: Annotated[StreamLogRequest, StreamLogRequest],
+                config_hash: str = "",
+            ) -> EventSourceResponse:
+                """Invoke the runnable stream_log the output.
+
+                This endpoint allows to stream the output of the runnable, including
+                the output of all intermediate steps.
+
+                The endpoint uses a server sent event stream to stream the output.
+
+                https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
+
+                Important: Set the "text/event-stream" media type for request headers if
+                           not using an existing SDK.
+
+                This endpoint uses two different types of events:
+
+                * data - for streaming the output of the runnable
+
+                    {
+                        "event": "data",
+                        "data": {
+                        ...
+                        }
+                    }
+
+                * error - for signaling an error in the stream, also ends the stream.
+
+                {
+                    "event": "error",
+                    "data": {
+                        "status_code": 500,
+                        "message": "Internal Server Error"
+                    }
+                }
+
+                * end - for signaling the end of the stream.
+
+                    This helps the client to know when to stop listening for events and
+                    know that the streaming has ended successfully.
+
+                    {
+                        "event": "end",
+                    }
+                """
+                raise AssertionError("This endpoint should not be reachable.")
+
+            app.post(
+                f"{namespace}/stream_log",
+                include_in_schema=True,
+                tags=route_tags,
+                name=_route_name("stream_log"),
                 dependencies=dependencies,
             )(_stream_log_docs)
 
-    if has_astream_events and endpoint_configuration.is_stream_events_enabled:
+            if endpoint_configuration.is_config_hash_enabled:
+                app.post(
+                    namespace + "/c/{config_hash}/stream_log",
+                    include_in_schema=True,
+                    tags=route_tags_with_config,
+                    name=_route_name_with_config("stream_log"),
+                    description=(
+                        "This endpoint is to be used with share links generated by the "
+                        "LangServe playground. "
+                        "The hash is an LZString compressed JSON string. "
+                        "For regular use cases, use the /stream_log endpoint without "
+                        "the `c/{config_hash}` path parameter."
+                    ),
+                    dependencies=dependencies,
+                )(_stream_log_docs)
 
-        async def _stream_events_docs(
-            stream_events_request: Annotated[StreamEventsRequest, StreamEventsRequest],
-            config_hash: str = "",
-        ) -> EventSourceResponse:
-            """Stream events from the given runnable.
+        if has_astream_events and endpoint_configuration.is_stream_events_enabled:
 
-            This endpoint allows to stream events from the runnable, including
-            events from all intermediate steps.
+            async def _stream_events_docs(
+                stream_events_request: Annotated[
+                    StreamEventsRequest, StreamEventsRequest
+                ],
+                config_hash: str = "",
+            ) -> EventSourceResponse:
+                """Stream events from the given runnable.
 
-            **Attention**
+                This endpoint allows to stream events from the runnable, including
+                events from all intermediate steps.
 
-                This is a new endpoint that only works for langchain-core >= 0.1.14.
+                **Attention**
 
-                It belongs to a Beta API that may change in the future.
+                    This is a new endpoint that only works for langchain-core >= 0.1.14.
 
-            **Important**
-                Specify filters to the events you want to receive by setting
-                the appropriate filters in the request body.
+                    It belongs to a Beta API that may change in the future.
 
-                This will help avoid sending too much data over the network.
+                **Important**
+                    Specify filters to the events you want to receive by setting
+                    the appropriate filters in the request body.
 
-                It will also prevent serialization issues with
-                any unsupported types since it won't need to serialize events
-                that aren't transmitted.
+                    This will help avoid sending too much data over the network.
 
-            The endpoint uses a server sent event stream to stream the output.
+                    It will also prevent serialization issues with
+                    any unsupported types since it won't need to serialize events
+                    that aren't transmitted.
 
-            https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
+                The endpoint uses a server sent event stream to stream the output.
 
-            The encoding of events follows the following format:
+                https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events
 
-            * data - for streaming the output of the runnable
+                The encoding of events follows the following format:
+
+                * data - for streaming the output of the runnable
+
+                    {
+                        "event": "data",
+                        "data": {
+                        ...
+                        }
+                    }
+
+                * error - for signaling an error in the stream, also ends the stream.
 
                 {
-                    "event": "data",
+                    "event": "error",
                     "data": {
-                    ...
+                        "status_code": 500,
+                        "message": "Internal Server Error"
                     }
                 }
 
-            * error - for signaling an error in the stream, also ends the stream.
+                * end - for signaling the end of the stream.
 
-            {
-                "event": "error",
-                "data": {
-                    "status_code": 500,
-                    "message": "Internal Server Error"
-                }
-            }
+                    This helps the client to know when to stop listening for events and
+                    know that the streaming has ended successfully.
 
-            * end - for signaling the end of the stream.
+                    {
+                        "event": "end",
+                    }
 
-                This helps the client to know when to stop listening for events and
-                know that the streaming has ended successfully.
+                `data` for the `data` event is a JSON object that corresponds
+                to a serialized representation of a StreamEvent.
 
-                {
-                    "event": "end",
-                }
+                See LangChain documentation for more information about astream_events.
+                """
+                raise AssertionError("This endpoint should not be reachable.")
 
-            `data` for the `data` event is a JSON object that corresponds
-            to a serialized representation of a StreamEvent.
-
-            See LangChain documentation for more information about astream_events.
-            """
-            raise AssertionError("This endpoint should not be reachable.")
-
-        app.post(
-            f"{namespace}/stream_events",
-            include_in_schema=True,
-            tags=route_tags,
-            name=_route_name("stream_events"),
-            dependencies=dependencies,
-        )(_stream_events_docs)
-
-        if endpoint_configuration.is_config_hash_enabled:
             app.post(
-                namespace + "/c/{config_hash}/stream_events",
+                f"{namespace}/stream_events",
                 include_in_schema=True,
-                tags=route_tags_with_config,
-                name=_route_name_with_config("stream_events"),
-                description=(
-                    "This endpoint is to be used with share links generated by the "
-                    "LangServe playground. "
-                    "The hash is an LZString compressed JSON string. "
-                    "For regular use cases, use the /stream_events endpoint "
-                    "without the `c/{config_hash}` path parameter."
-                ),
+                tags=route_tags,
+                name=_route_name("stream_events"),
                 dependencies=dependencies,
             )(_stream_events_docs)
+
+            if endpoint_configuration.is_config_hash_enabled:
+                app.post(
+                    namespace + "/c/{config_hash}/stream_events",
+                    include_in_schema=True,
+                    tags=route_tags_with_config,
+                    name=_route_name_with_config("stream_events"),
+                    description=(
+                        "This endpoint is to be used with share links generated by the "
+                        "LangServe playground. "
+                        "The hash is an LZString compressed JSON string. "
+                        "For regular use cases, use the /stream_events endpoint "
+                        "without the `c/{config_hash}` path parameter."
+                    ),
+                    dependencies=dependencies,
+                )(_stream_events_docs)

langserve/server_sent_events.py

@@ -1,9 +1,8 @@
 """Adapted from https://github.com/florimondmanca/httpx-sse"""
 from contextlib import asynccontextmanager, contextmanager
-from typing import Any, AsyncIterator, Iterator, List, Optional
+from typing import Any, AsyncIterator, Iterator, List, Optional, TypedDict
 
 import httpx
-from typing_extensions import TypedDict
 
 
 class ServerSentEvent(TypedDict):

langserve/validation.py

@@ -22,11 +22,16 @@ from uuid import UUID
 from langchain_core.documents import Document
 from langchain_core.messages import BaseMessage
 from langchain_core.outputs import ChatGeneration, Generation, RunInfo
-from pydantic import BaseModel, Field, RootModel, create_model
 from typing_extensions import Type
 
 from langserve.schema import BatchResponseMetadata, InvokeResponseMetadata
 
+try:
+    from pydantic.v1 import BaseModel, Field, create_model
+except ImportError:
+    from pydantic import BaseModel, Field, create_model
+
+
 # Type that is either a python annotation or a pydantic model that can be
 # used to validate the input or output of a runnable.
 Validator = Union[Type[BaseModel], type]
@@ -61,7 +66,7 @@ def create_invoke_request_model(
             ),
         ),
     )
-    invoke_request_type.model_rebuild()
+    invoke_request_type.update_forward_refs()
     return invoke_request_type
 
 
@@ -92,7 +97,7 @@ def create_stream_request_model(
             ),
         ),
     )
-    stream_request_model.model_rebuild()
+    stream_request_model.update_forward_refs()
     return stream_request_model
 
 
@@ -124,7 +129,7 @@ def create_batch_request_model(
             ),
         ),
     )
-    batch_request_type.model_rebuild()
+    batch_request_type.update_forward_refs()
     return batch_request_type
 
 
@@ -182,7 +187,7 @@ def create_stream_log_request_model(
         ),
         kwargs=(dict, Field(default_factory=dict)),
     )
-    stream_log_request.model_rebuild()
+    stream_log_request.update_forward_refs()
     return stream_log_request
 
 
@@ -240,7 +245,7 @@ def create_stream_events_request_model(
         ),
         kwargs=(dict, Field(default_factory=dict)),
     )
-    stream_events_request.model_rebuild()
+    stream_events_request.update_forward_refs()
     return stream_events_request
 
 
@@ -292,7 +297,7 @@ def create_invoke_response_model(
         __base__=InvokeBaseResponse,
         **fields,
     )
-    invoke_response_type.model_rebuild()
+    invoke_response_type.update_forward_refs()
     return invoke_response_type
 
 
@@ -342,7 +347,7 @@ def create_batch_response_model(
         __base__=BatchBaseResponse,
         **fields,
     )
-    batch_response_type.model_rebuild()
+    batch_response_type.update_forward_refs()
     return batch_response_type
 
 
@@ -396,29 +401,27 @@ class StreamEventsParameters(BaseModel):
 # status code and a message.
 
 
-class BaseCallback(BaseModel):
-    """Base class for all callback events."""
+class OnChainStart(BaseModel):
+    """On Chain Start Callback Event."""
 
+    serialized: Dict[str, Any]
+    inputs: Any
     run_id: UUID
     parent_run_id: Optional[UUID] = None
     tags: Optional[List[str]] = None
     metadata: Optional[Dict[str, Any]] = None
-
-
-class OnChainStart(BaseCallback):
-    """On Chain Start Callback Event."""
-
-    serialized: Optional[Dict[str, Any]] = None
-    inputs: Any
-    kwargs: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_chain_start"] = "on_chain_start"
 
 
-class OnChainEnd(BaseCallback):
+class OnChainEnd(BaseModel):
     """On Chain End Callback Event."""
 
     outputs: Any
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    kwargs: Any = None
     type: Literal["on_chain_end"] = "on_chain_end"
 
 
@@ -430,35 +433,38 @@ class Error(BaseModel):
     type: Literal["error"] = "error"
 
 
-class OnChainError(BaseCallback):
+class OnChainError(BaseModel):
     """On Chain Error Callback Event."""
 
     error: Error
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    kwargs: Any = None
     type: Literal["on_chain_error"] = "on_chain_error"
 
 
-class OnToolStart(BaseCallback):
+class OnToolStart(BaseModel):
     """On Tool Start Callback Event."""
 
-    serialized: Optional[Dict[str, Any]] = None
+    serialized: Dict[str, Any]
     input_str: str
     run_id: UUID
     parent_run_id: Optional[UUID] = None
     tags: Optional[List[str]] = None
     metadata: Optional[Dict[str, Any]] = None
-    kwargs: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_tool_start"] = "on_tool_start"
 
 
-class OnToolEnd(BaseCallback):
+class OnToolEnd(BaseModel):
     """On Tool End Callback Event."""
 
     output: str
     run_id: UUID
     parent_run_id: Optional[UUID] = None
     tags: Optional[List[str]] = None
-    kwargs: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_tool_end"] = "on_tool_end"
 
 
@@ -466,29 +472,36 @@ class OnToolError(BaseModel):
     """On Tool Error Callback Event."""
 
     error: Error
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    kwargs: Any = None
     type: Literal["on_tool_error"] = "on_tool_error"
 
 
-class OnChatModelStart(BaseCallback):
+class OnChatModelStart(BaseModel):
     """On Chat Model Start Callback Event."""
 
-    serialized: Optional[Dict[str, Any]] = None
+    serialized: Dict[str, Any]
     messages: List[List[BaseMessage]]
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    metadata: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_chat_model_start"] = "on_chat_model_start"
 
 
-class OnLLMStart(BaseCallback):
+class OnLLMStart(BaseModel):
     """On LLM Start Callback Event."""
 
-    serialized: Optional[Dict[str, Any]] = None
+    serialized: Dict[str, Any]
     prompts: List[str]
     run_id: UUID
     parent_run_id: Optional[UUID] = None
     tags: Optional[List[str]] = None
     metadata: Optional[Dict[str, Any]] = None
-    kwargs: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_llm_start"] = "on_llm_start"
 
 
@@ -507,44 +520,54 @@ class LLMResult(BaseModel):
     """List of metadata info for model call for each input."""
 
 
-class OnLLMEnd(BaseCallback):
+class OnLLMEnd(BaseModel):
     """On LLM End Callback Event."""
 
     response: LLMResult
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    kwargs: Any = None
     type: Literal["on_llm_end"] = "on_llm_end"
 
 
-class OnRetrieverStart(BaseCallback):
+class OnRetrieverStart(BaseModel):
     """On Retriever Start Callback Event."""
 
-    serialized: Optional[Dict[str, Any]] = None
+    serialized: Dict[str, Any]
     query: str
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    metadata: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_retriever_start"] = "on_retriever_start"
 
 
-class OnRetrieverError(BaseCallback):
+class OnRetrieverError(BaseModel):
     """On Retriever Error Callback Event."""
 
     error: Error
     run_id: UUID
     parent_run_id: Optional[UUID] = None
     tags: Optional[List[str]] = None
-    kwargs: Optional[Dict[str, Any]] = None
+    kwargs: Any = None
     type: Literal["on_retriever_error"] = "on_retriever_error"
 
 
-class OnRetrieverEnd(BaseCallback):
+class OnRetrieverEnd(BaseModel):
     """On Retriever End Callback Event."""
 
     documents: Sequence[Document]
-    kwargs: Optional[Dict[str, Any]] = None
+    run_id: UUID
+    parent_run_id: Optional[UUID] = None
+    tags: Optional[List[str]] = None
+    kwargs: Any = None
     type: Literal["on_retriever_end"] = "on_retriever_end"
 
 
-CallbackEvent = RootModel[
-    Union[
+class CallbackEvent(BaseModel):
+    __root__: Union[
         OnChainStart,
         OnChainEnd,
         OnChainError,
@@ -558,4 +581,3 @@ CallbackEvent = RootModel[
         OnRetrieverEnd,
         OnRetrieverError,
     ]
-]

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "langserve"
-version = "0.3.0"
+version = "0.2.1"
 description = ""
 readme = "README.md"
 authors = ["LangChain"]
@@ -10,13 +10,14 @@ exclude = ["langserve/playground,langserve/chat_playground"]
 include = ["langserve/playground/dist/**/*", "langserve/chat_playground/dist/**/*"]
 
 [tool.poetry.dependencies]
-python = "^3.9"
-httpx = ">=0.23.0,<1.0"
+python = "^3.8.1"
+httpx = ">=0.23.0" # May be able to decrease this version
 fastapi = {version = ">=0.90.1,<1", optional = true}
 sse-starlette = {version = "^1.3.0", optional = true}
-langchain-core = "^0.3"
-orjson = ">=2,<4"
-pydantic = "^2.7"
+pydantic = ">=1"
+langchain-core = ">=0.1,<0.3"
+orjson = ">=2"
+pyproject-toml = "^0.0.10"
 
 [tool.poetry.group.dev.dependencies]
 jupyterlab = "^3.6.1"
@@ -46,7 +47,7 @@ sse-starlette = "^1.3.0"
 
 [tool.poetry.extras]
 # Extras that are used for client
-client = ["fastapi"]
+client = []
 # Extras that are used for server
 server = ["sse-starlette", "fastapi"]
 # All
@@ -94,7 +95,3 @@ addopts = "--strict-markers --strict-config --durations=5 -vv"
 # take more than 5 seconds
 timeout = 5
 asyncio_mode = "auto"
-filterwarnings =  [
-    "ignore::langchain_core._api.beta_decorator.LangChainBetaWarning",
-]
-

tests/unit_tests/test_api_playground.py

@@ -1,6 +1,5 @@
 """Test the playground API."""
 
-import httpx
 from fastapi import APIRouter, FastAPI
 from httpx import AsyncClient
 from langchain_core.runnables import RunnableLambda
@@ -16,9 +15,7 @@ async def test_serve_playground() -> None:
         RunnableLambda(lambda foo: "hello"),
     )
 
-    async with AsyncClient(
-        base_url="http://localhost:9999", transport=httpx.ASGITransport(app=app)
-    ) as client:
+    async with AsyncClient(app=app, base_url="http://localhost:9999") as client:
         response = await client.get("/playground/index.html")
         assert response.status_code == 200
         # Test that we can't access files that do not exist.
@@ -45,9 +42,7 @@ async def test_serve_playground_with_api_router() -> None:
 
     app.include_router(router)
 
-    async with AsyncClient(
-        base_url="http://localhost:9999", transport=httpx.ASGITransport(app=app)
-    ) as client:
+    async with AsyncClient(app=app, base_url="http://localhost:9999") as client:
         response = await client.get("/langserve_runnables/chat/playground/index.html")
         assert response.status_code == 200
 
@@ -69,9 +64,7 @@ async def test_serve_playground_with_api_router_in_api_router() -> None:
     # Now add parent router to the app
     app.include_router(parent_router)
 
-    async with AsyncClient(
-        base_url="http://localhost:9999", transport=httpx.ASGITransport(app=app)
-    ) as client:
+    async with AsyncClient(app=app, base_url="http://localhost:9999") as client:
         response = await client.get("/parent/bar/foo/playground/index.html")
         assert response.status_code == 200
 
@@ -95,9 +88,7 @@ async def test_root_path_on_playground() -> None:
         )
         app.include_router(router)
 
-        async_client = AsyncClient(
-            base_url="http://localhost:9999", transport=httpx.ASGITransport(app=app)
-        )
+        async_client = AsyncClient(app=app, base_url="http://localhost:9999")
 
         response = await async_client.get("/chat/playground/index.html")
         assert response.status_code == 200

tests/unit_tests/test_serialization.py

@@ -4,23 +4,15 @@ from enum import Enum
 from typing import Any
 
 import pytest
-from langchain_core.documents.base import Document
 from langchain_core.messages import HumanMessage, HumanMessageChunk, SystemMessage
 from langchain_core.outputs import ChatGeneration
-from pydantic import BaseModel
 
-from langserve.serialization import (
-    WellKnownLCObject,
-    WellKnownLCSerializer,
-    load_events,
-)
+try:
+    from pydantic.v1 import BaseModel
+except ImportError:
+    from pydantic import BaseModel
 
-
-def test_document_serialization() -> None:
-    """Simple test. Exhaustive tests follow below."""
-    doc = Document(page_content="hello")
-    d = doc.model_dump()
-    WellKnownLCObject.model_validate(d)
+from langserve.serialization import WellKnownLCSerializer, load_events
 
 
 @pytest.mark.parametrize(
@@ -31,8 +23,6 @@ def test_document_serialization() -> None:
         [],
         {},
         {"a": 1},
-        Document(page_content="Hello"),
-        [Document(page_content="Hello")],
         {"output": [HumanMessage(content="hello")]},
         # Test with a single message (HumanMessage)
         HumanMessage(content="Hello"),
@@ -87,7 +77,7 @@ def _get_full_representation(data: Any) -> Any:
     elif isinstance(data, list):
         return [_get_full_representation(value) for value in data]
     elif isinstance(data, BaseModel):
-        return data.model_json_schema()
+        return data.schema()
     else:
         return data
 

tests/unit_tests/test_validation.py

@@ -5,9 +5,14 @@ import pytest
 from fastapi import Request
 from langchain_core.prompts import PromptTemplate
 from langchain_core.runnables import ConfigurableField
-from pydantic import BaseModel, ValidationError
 
 from langserve.api_handler import _unpack_request_config
+
+try:
+    from pydantic.v1 import BaseModel, ValidationError
+except ImportError:
+    from pydantic import BaseModel, ValidationError
+
 from langserve.validation import (
     create_batch_request_model,
     create_invoke_request_model,
@@ -170,11 +175,11 @@ async def test_invoke_request_with_runnables() -> None:
             "configurable": {"template": "goodbye {name}"},
         },
     )
-    assert dict(request.input) == {"name": "bob"}
+    assert request.input == {"name": "bob"}
     assert request.config.tags == ["hello"]
     assert request.config.run_name == "run"
     assert isinstance(request.config.configurable, BaseModel)
-    assert request.config.configurable.model_dump() == {
+    assert request.config.configurable.dict() == {
         "template": "goodbye {name}",
     }
 

tests/unit_tests/utils/serde.py

@@ -1,27 +0,0 @@
-from typing import Any
-
-
-def recursive_dump(obj: Any) -> Any:
-    """Recursively dump the object if encountering any pydantic models."""
-    if isinstance(obj, dict):
-        return {
-            k: recursive_dump(v)
-            for k, v in obj.items()
-            if k != "id"  # Remove the id field for testing purposes
-        }
-    if isinstance(obj, list):
-        return [recursive_dump(v) for v in obj]
-    if hasattr(obj, "model_dump"):
-        # if the object contains an ID field, we'll remove it for testing purposes
-        d = obj.model_dump()
-        if "id" in d:
-            d.pop("id")
-        return recursive_dump(d)
-    if hasattr(obj, "dict"):
-        # if the object contains an ID field, we'll remove it for testing purposes
-        if hasattr(obj, "id"):
-            d = obj.dict()
-            d.pop("id")
-            return recursive_dump(d)
-        return recursive_dump(obj.dict())
-    return obj

tests/unit_tests/utils/stubs.py

@@ -1,22 +1,6 @@
 from typing import Any
 
-from langchain_core.messages import AIMessage, AIMessageChunk
-
 
 class AnyStr(str):
     def __eq__(self, other: Any) -> bool:
         return isinstance(other, str)
-
-
-def _AnyIdAIMessage(**kwargs: Any) -> AIMessage:
-    """Create ai message with an any id field."""
-    message = AIMessage(**kwargs)
-    message.id = AnyStr()
-    return message
-
-
-def _AnyIdAIMessageChunk(**kwargs: Any) -> AIMessageChunk:
-    """Create ai message with an any id field."""
-    message = AIMessageChunk(**kwargs)
-    message.id = AnyStr()
-    return message

tests/unit_tests/utils/test_fake_chat_model.py

@@ -4,11 +4,11 @@ from typing import Any, Dict, List, Optional, Union
 from uuid import UUID
 
 from langchain_core.callbacks.base import AsyncCallbackHandler
-from langchain_core.messages import AIMessage, BaseMessage
+from langchain_core.messages import AIMessage, AIMessageChunk, BaseMessage
 from langchain_core.outputs import ChatGenerationChunk, GenerationChunk
 
 from tests.unit_tests.utils.llms import GenericFakeChatModel
-from tests.unit_tests.utils.stubs import _AnyIdAIMessage, _AnyIdAIMessageChunk
+from tests.unit_tests.utils.stubs import AnyStr
 
 
 def test_generic_fake_chat_model_invoke() -> None:
@@ -16,11 +16,11 @@ def test_generic_fake_chat_model_invoke() -> None:
     infinite_cycle = cycle([AIMessage(content="hello"), AIMessage(content="goodbye")])
     model = GenericFakeChatModel(messages=infinite_cycle)
     response = model.invoke("meow")
-    assert response == _AnyIdAIMessage(content="hello")
+    assert response == AIMessage(content="hello", id=AnyStr())
     response = model.invoke("kitty")
-    assert response == _AnyIdAIMessage(content="goodbye")
+    assert response == AIMessage(content="goodbye", id=AnyStr())
     response = model.invoke("meow")
-    assert response == _AnyIdAIMessage(content="hello")
+    assert response == AIMessage(content="hello", id=AnyStr())
 
 
 async def test_generic_fake_chat_model_ainvoke() -> None:
@@ -28,11 +28,11 @@ async def test_generic_fake_chat_model_ainvoke() -> None:
     infinite_cycle = cycle([AIMessage(content="hello"), AIMessage(content="goodbye")])
     model = GenericFakeChatModel(messages=infinite_cycle)
     response = await model.ainvoke("meow")
-    assert response == _AnyIdAIMessage(content="hello")
+    assert response == AIMessage(content="hello", id=AnyStr())
     response = await model.ainvoke("kitty")
-    assert response == _AnyIdAIMessage(content="goodbye")
+    assert response == AIMessage(content="goodbye", id=AnyStr())
     response = await model.ainvoke("meow")
-    assert response == _AnyIdAIMessage(content="hello")
+    assert response == AIMessage(content="hello", id=AnyStr())
 
 
 async def test_generic_fake_chat_model_stream() -> None:
@@ -45,26 +45,28 @@ async def test_generic_fake_chat_model_stream() -> None:
     model = GenericFakeChatModel(messages=infinite_cycle)
     chunks = [chunk async for chunk in model.astream("meow")]
     assert chunks == [
-        _AnyIdAIMessageChunk(content="hello"),
-        _AnyIdAIMessageChunk(content=" "),
-        _AnyIdAIMessageChunk(content="goodbye"),
+        AIMessageChunk(content="hello", id=AnyStr()),
+        AIMessageChunk(content=" ", id=AnyStr()),
+        AIMessageChunk(content="goodbye", id=AnyStr()),
     ]
 
     chunks = [chunk for chunk in model.stream("meow")]
     assert chunks == [
-        _AnyIdAIMessageChunk(content="hello"),
-        _AnyIdAIMessageChunk(content=" "),
-        _AnyIdAIMessageChunk(content="goodbye"),
+        AIMessageChunk(content="hello", id=AnyStr()),
+        AIMessageChunk(content=" ", id=AnyStr()),
+        AIMessageChunk(content="goodbye", id=AnyStr()),
     ]
 
     # Test streaming of additional kwargs.
     # Relying on insertion order of the additional kwargs dict
-    message = AIMessage(content="", additional_kwargs={"foo": 42, "bar": 24}, id="1")
+    message = AIMessage(
+        content="", additional_kwargs={"foo": 42, "bar": 24}, id=AnyStr()
+    )
     model = GenericFakeChatModel(messages=cycle([message]))
     chunks = [chunk async for chunk in model.astream("meow")]
     assert chunks == [
-        _AnyIdAIMessageChunk(content="", additional_kwargs={"foo": 42}),
-        _AnyIdAIMessageChunk(content="", additional_kwargs={"bar": 24}),
+        AIMessageChunk(content="", additional_kwargs={"foo": 42}, id=AnyStr()),
+        AIMessageChunk(content="", additional_kwargs={"bar": 24}, id=AnyStr()),
     ]
 
     message = AIMessage(
@@ -81,25 +83,29 @@ async def test_generic_fake_chat_model_stream() -> None:
     chunks = [chunk async for chunk in model.astream("meow")]
 
     assert chunks == [
-        _AnyIdAIMessageChunk(
+        AIMessageChunk(
             content="",
             additional_kwargs={"function_call": {"name": "move_file"}},
+            id=AnyStr(),
         ),
-        _AnyIdAIMessageChunk(
+        AIMessageChunk(
             content="",
             additional_kwargs={
                 "function_call": {"arguments": '{\n  "source_path": "foo"'}
             },
+            id=AnyStr(),
         ),
-        _AnyIdAIMessageChunk(
+        AIMessageChunk(
             content="",
             additional_kwargs={"function_call": {"arguments": ","}},
+            id=AnyStr(),
         ),
-        _AnyIdAIMessageChunk(
+        AIMessageChunk(
             content="",
             additional_kwargs={
                 "function_call": {"arguments": '\n  "destination_path": "bar"\n}'}
             },
+            id=AnyStr(),
         ),
     ]
 
@@ -110,7 +116,7 @@ async def test_generic_fake_chat_model_stream() -> None:
         else:
             accumulate_chunks += chunk
 
-    assert accumulate_chunks == _AnyIdAIMessageChunk(
+    assert accumulate_chunks == AIMessageChunk(
         content="",
         additional_kwargs={
             "function_call": {
@@ -119,6 +125,7 @@ async def test_generic_fake_chat_model_stream() -> None:
                 'destination_path": "bar"\n}',
             }
         },
+        id=AnyStr(),
     )
 
 
@@ -131,9 +138,9 @@ async def test_generic_fake_chat_model_astream_log() -> None:
     ]
     final = log_patches[-1]
     assert final.state["streamed_output"] == [
-        _AnyIdAIMessageChunk(content="hello"),
-        _AnyIdAIMessageChunk(content=" "),
-        _AnyIdAIMessageChunk(content="goodbye"),
+        AIMessageChunk(content="hello", id=AnyStr()),
+        AIMessageChunk(content=" ", id=AnyStr()),
+        AIMessageChunk(content="goodbye", id=AnyStr()),
     ]
 
 
@@ -181,8 +188,8 @@ async def test_callback_handlers() -> None:
     # New model
     results = list(model.stream("meow", {"callbacks": [MyCustomAsyncHandler(tokens)]}))
     assert results == [
-        _AnyIdAIMessageChunk(content="hello"),
-        _AnyIdAIMessageChunk(content=" "),
-        _AnyIdAIMessageChunk(content="goodbye"),
+        AIMessageChunk(content="hello", id=AnyStr()),
+        AIMessageChunk(content=" ", id=AnyStr()),
+        AIMessageChunk(content="goodbye", id=AnyStr()),
     ]
     assert tokens == ["hello", " ", "goodbye"]

tests/unit_tests/utils/tracer.py

@@ -1,4 +1,4 @@
-from typing import Any, Dict, List, Optional
+from typing import Dict, List
 from uuid import UUID
 
 from langchain_core.tracers import BaseTracer
@@ -39,34 +39,6 @@ class FakeTracer(BaseTracer):
             }
         )
 
-    def _create_chain_run(
-        self,
-        serialized: Dict[str, Any],
-        inputs: Dict[str, Any],
-        run_id: UUID,
-        tags: Optional[List[str]] = None,
-        parent_run_id: Optional[UUID] = None,
-        metadata: Optional[Dict[str, Any]] = None,
-        run_type: Optional[str] = None,
-        name: Optional[str] = None,
-        **kwargs: Any,
-    ) -> Run:
-        if name is None:
-            # can't raise an exception from here, but can get a breakpoint
-            # import pdb; pdb.set_trace()
-            pass
-        return super()._create_chain_run(
-            serialized,
-            inputs,
-            run_id,
-            tags,
-            parent_run_id,
-            metadata,
-            run_type,
-            name,
-            **kwargs,
-        )
-
     def _persist_run(self, run: Run) -> None:
         """Persist a run."""
         self.runs.append(self._copy_run(run))