langchain-core 0.4.0.dev0__py3-none-any.whl → 1.0.0__py3-none-any.whl

langchain_core/language_models/chat_models.py

@@ -6,28 +6,28 @@ import asyncio
 import inspect
 import json
 import typing
-import warnings
 from abc import ABC, abstractmethod
-from collections.abc import AsyncIterator, Iterator, Sequence
+from collections.abc import AsyncIterator, Callable, Iterator, Sequence
 from functools import cached_property
 from operator import itemgetter
-from typing import TYPE_CHECKING, Any, Callable, Literal, Optional, Union, cast
+from typing import TYPE_CHECKING, Any, Literal, cast
 
-from pydantic import BaseModel, ConfigDict, Field, model_validator
+from pydantic import BaseModel, ConfigDict, Field
 from typing_extensions import override
 
-from langchain_core._api import deprecated
 from langchain_core.caches import BaseCache
 from langchain_core.callbacks import (
     AsyncCallbackManager,
     AsyncCallbackManagerForLLMRun,
-    BaseCallbackManager,
     CallbackManager,
     CallbackManagerForLLMRun,
     Callbacks,
 )
 from langchain_core.globals import get_llm_cache
-from langchain_core.language_models._utils import _normalize_messages
+from langchain_core.language_models._utils import (
+    _normalize_messages,
+    _update_message_content_to_blocks,
+)
 from langchain_core.language_models.base import (
     BaseLanguageModel,
     LangSmithParams,
@@ -36,16 +36,21 @@ from langchain_core.language_models.base import (
 from langchain_core.load import dumpd, dumps
 from langchain_core.messages import (
     AIMessage,
+    AIMessageChunk,
     AnyMessage,
     BaseMessage,
-    BaseMessageChunk,
-    HumanMessage,
     convert_to_messages,
-    convert_to_openai_image_block,
     is_data_content_block,
     message_chunk_to_message,
 )
-from langchain_core.messages.ai import _LC_ID_PREFIX
+from langchain_core.messages import content as types
+from langchain_core.messages.block_translators.openai import (
+    convert_to_openai_image_block,
+)
+from langchain_core.output_parsers.openai_tools import (
+    JsonOutputKeyToolsParser,
+    PydanticToolsParser,
+)
 from langchain_core.outputs import (
     ChatGeneration,
     ChatGenerationChunk,
@@ -65,6 +70,7 @@ from langchain_core.utils.function_calling import (
     convert_to_openai_tool,
 )
 from langchain_core.utils.pydantic import TypeBaseModel, is_basemodel_subclass
+from langchain_core.utils.utils import LC_ID_PREFIX, from_env
 
 if TYPE_CHECKING:
     import uuid
@@ -78,6 +84,11 @@ def _generate_response_from_error(error: BaseException) -> list[ChatGeneration]:
     if hasattr(error, "response"):
         response = error.response
         metadata: dict = {}
+        if hasattr(response, "json"):
+            try:
+                metadata["body"] = response.json()
+            except Exception:
+                metadata["body"] = getattr(response, "text", None)
         if hasattr(response, "headers"):
             try:
                 metadata["headers"] = dict(response.headers)
@@ -97,17 +108,18 @@ def _generate_response_from_error(error: BaseException) -> list[ChatGeneration]:
 
 
 def _format_for_tracing(messages: list[BaseMessage]) -> list[BaseMessage]:
-    """Format messages for tracing in on_chat_model_start.
+    """Format messages for tracing in `on_chat_model_start`.
 
     - Update image content blocks to OpenAI Chat Completions format (backward
     compatibility).
-    - Add "type" key to content blocks that have a single key.
+    - Add `type` key to content blocks that have a single key.
 
     Args:
         messages: List of messages to format.
 
     Returns:
         List of messages formatted for tracing.
+
     """
     messages_to_trace = []
     for message in messages:
@@ -119,7 +131,7 @@ def _format_for_tracing(messages: list[BaseMessage]) -> list[BaseMessage]:
                     if (
                         block.get("type") == "image"
                         and is_data_content_block(block)
-                        and block.get("source_type") != "id"
+                        and not ("file_id" in block or block.get("source_type") == "id")
                     ):
                         if message_to_trace is message:
                             # Shallow copy
@@ -129,6 +141,22 @@ def _format_for_tracing(messages: list[BaseMessage]) -> list[BaseMessage]:
                         message_to_trace.content[idx] = (  # type: ignore[index]  # mypy confused by .model_copy
                             convert_to_openai_image_block(block)
                         )
+                    elif (
+                        block.get("type") == "file"
+                        and is_data_content_block(block)  # v0 (image/audio/file) or v1
+                        and "base64" in block
+                        # Backward compat: convert v1 base64 blocks to v0
+                    ):
+                        if message_to_trace is message:
+                            # Shallow copy
+                            message_to_trace = message.model_copy()
+                            message_to_trace.content = list(message_to_trace.content)
+
+                        message_to_trace.content[idx] = {  # type: ignore[index]
+                            **{k: v for k, v in block.items() if k != "base64"},
+                            "data": block["base64"],
+                            "source_type": "base64",
+                        }
                     elif len(block) == 1 and "type" not in block:
                         # Tracing assumes all content blocks have a "type" key. Here
                         # we add this key if it is missing, and there's an obvious
@@ -142,8 +170,6 @@ def _format_for_tracing(messages: list[BaseMessage]) -> list[BaseMessage]:
                             "type": key,
                             key: block[key],
                         }
-                    else:
-                        pass
         messages_to_trace.append(message_to_trace)
 
     return messages_to_trace
@@ -153,10 +179,14 @@ def generate_from_stream(stream: Iterator[ChatGenerationChunk]) -> ChatResult:
     """Generate from a stream.
 
     Args:
-        stream: Iterator of ChatGenerationChunk.
+        stream: Iterator of `ChatGenerationChunk`.
+
+    Raises:
+        ValueError: If no generations are found in the stream.
 
     Returns:
-        ChatResult: Chat result.
+        Chat result.
+
     """
     generation = next(stream, None)
     if generation:
@@ -180,16 +210,17 @@ async def agenerate_from_stream(
     """Async generate from a stream.
 
     Args:
-        stream: Iterator of ChatGenerationChunk.
+        stream: Iterator of `ChatGenerationChunk`.
 
     Returns:
-        ChatResult: Chat result.
+        Chat result.
+
     """
     chunks = [chunk async for chunk in stream]
     return await run_in_executor(None, generate_from_stream, iter(chunks))
 
 
-def _format_ls_structured_output(ls_structured_output_format: Optional[dict]) -> dict:
+def _format_ls_structured_output(ls_structured_output_format: dict | None) -> dict:
     if ls_structured_output_format:
         try:
             ls_structured_output_format_dict = {
@@ -208,136 +239,99 @@ def _format_ls_structured_output(ls_structured_output_format: Optional[dict]) ->
     return ls_structured_output_format_dict
 
 
-class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
-    """Base class for chat models.
+class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
+    r"""Base class for chat models.
 
     Key imperative methods:
         Methods that actually call the underlying model.
 
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | Method                    | Input                                                          | Output                                                              | Description                                                                                      |
-        +===========================+================================================================+=====================================================================+==================================================================================================+
-        | `invoke`                  | str | list[dict | tuple | BaseMessage] | PromptValue           | BaseMessage                                                         | A single chat model call.                                                                        |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `ainvoke`                 | '''                                                            | BaseMessage                                                         | Defaults to running invoke in an async executor.                                                 |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `stream`                  | '''                                                            | Iterator[BaseMessageChunk]                                          | Defaults to yielding output of invoke.                                                           |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `astream`                 | '''                                                            | AsyncIterator[BaseMessageChunk]                                     | Defaults to yielding output of ainvoke.                                                          |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `astream_events`          | '''                                                            | AsyncIterator[StreamEvent]                                          | Event types: 'on_chat_model_start', 'on_chat_model_stream', 'on_chat_model_end'.                 |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `batch`                   | list[''']                                                      | list[BaseMessage]                                                   | Defaults to running invoke in concurrent threads.                                                |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `abatch`                  | list[''']                                                      | list[BaseMessage]                                                   | Defaults to running ainvoke in concurrent threads.                                               |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `batch_as_completed`      | list[''']                                                      | Iterator[tuple[int, Union[BaseMessage, Exception]]]                 | Defaults to running invoke in concurrent threads.                                                |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-        | `abatch_as_completed`     | list[''']                                                      | AsyncIterator[tuple[int, Union[BaseMessage, Exception]]]            | Defaults to running ainvoke in concurrent threads.                                               |
-        +---------------------------+----------------------------------------------------------------+---------------------------------------------------------------------+--------------------------------------------------------------------------------------------------+
-
-        This table provides a brief overview of the main imperative methods. Please see the base Runnable reference for full documentation.
+        This table provides a brief overview of the main imperative methods. Please see the base `Runnable` reference for full documentation.
+
+        | Method                 | Input                                                        | Output                                                     | Description                                                                      |
+        | ---------------------- | ------------------------------------------------------------ | ---------------------------------------------------------- | -------------------------------------------------------------------------------- |
+        | `invoke`               | `str` \| `list[dict | tuple | BaseMessage]` \| `PromptValue` | `BaseMessage`                                              | A single chat model call.                                                        |
+        | `ainvoke`              | `'''`                                                        | `BaseMessage`                                              | Defaults to running `invoke` in an async executor.                               |
+        | `stream`               | `'''`                                                        | `Iterator[BaseMessageChunk]`                               | Defaults to yielding output of `invoke`.                                         |
+        | `astream`              | `'''`                                                        | `AsyncIterator[BaseMessageChunk]`                          | Defaults to yielding output of `ainvoke`.                                        |
+        | `astream_events`       | `'''`                                                        | `AsyncIterator[StreamEvent]`                               | Event types: `on_chat_model_start`, `on_chat_model_stream`, `on_chat_model_end`. |
+        | `batch`                | `list[''']`                                                  | `list[BaseMessage]`                                        | Defaults to running `invoke` in concurrent threads.                              |
+        | `abatch`               | `list[''']`                                                  | `list[BaseMessage]`                                        | Defaults to running `ainvoke` in concurrent threads.                             |
+        | `batch_as_completed`   | `list[''']`                                                  | `Iterator[tuple[int, Union[BaseMessage, Exception]]]`      | Defaults to running `invoke` in concurrent threads.                              |
+        | `abatch_as_completed`  | `list[''']`                                                  | `AsyncIterator[tuple[int, Union[BaseMessage, Exception]]]` | Defaults to running `ainvoke` in concurrent threads.                             |
 
     Key declarative methods:
-        Methods for creating another Runnable using the ChatModel.
-
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | Method                           | Description                                                                                               |
-        +==================================+===========================================================================================================+
-        | `bind_tools`                     | Create ChatModel that can call tools.                                                                     |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | `with_structured_output`         | Create wrapper that structures model output using schema.                                                 |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | `with_retry`                     | Create wrapper that retries model calls on failure.                                                       |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | `with_fallbacks`                 | Create wrapper that falls back to other models on failure.                                                |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | `configurable_fields`            | Specify init args of the model that can be configured at runtime via the RunnableConfig.                  |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
-        | `configurable_alternatives`      | Specify alternative models which can be swapped in at runtime via the RunnableConfig.                     |
-        +----------------------------------+-----------------------------------------------------------------------------------------------------------+
+        Methods for creating another `Runnable` using the chat model.
 
         This table provides a brief overview of the main declarative methods. Please see the reference for each method for full documentation.
 
+        | Method                       | Description                                                                                |
+        | ---------------------------- | ------------------------------------------------------------------------------------------ |
+        | `bind_tools`                 | Create chat model that can call tools.                                                     |
+        | `with_structured_output`     | Create wrapper that structures model output using schema.                                  |
+        | `with_retry`                 | Create wrapper that retries model calls on failure.                                        |
+        | `with_fallbacks`             | Create wrapper that falls back to other models on failure.                                 |
+        | `configurable_fields`        | Specify init args of the model that can be configured at runtime via the `RunnableConfig`. |
+        | `configurable_alternatives`  | Specify alternative models which can be swapped in at runtime via the `RunnableConfig`.    |
+
     Creating custom chat model:
         Custom chat model implementations should inherit from this class.
         Please reference the table below for information about which
         methods and properties are required or optional for implementations.
 
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
-        | Method/Property                  | Description                                                        | Required/Optional |
-        +==================================+====================================================================+===================+
+        | Method/Property                  | Description                                                        | Required          |
+        | -------------------------------- | ------------------------------------------------------------------ | ----------------- |
         | `_generate`                      | Use to generate a chat result from a prompt                        | Required          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
         | `_llm_type` (property)           | Used to uniquely identify the type of the model. Used for logging. | Required          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
         | `_identifying_params` (property) | Represent model parameterization for tracing purposes.             | Optional          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
         | `_stream`                        | Use to implement streaming                                         | Optional          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
         | `_agenerate`                     | Use to implement a native async method                             | Optional          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
         | `_astream`                       | Use to implement async version of `_stream`                        | Optional          |
-        +----------------------------------+--------------------------------------------------------------------+-------------------+
-
-        Follow the guide for more information on how to implement a custom Chat Model:
-        [Guide](https://python.langchain.com/docs/how_to/custom_chat_model/).
 
     """  # noqa: E501
 
-    callback_manager: Optional[BaseCallbackManager] = deprecated(
-        name="callback_manager", since="0.1.7", removal="1.0", alternative="callbacks"
-    )(
-        Field(
-            default=None,
-            exclude=True,
-            description="Callback manager to add to the run trace.",
-        )
-    )
-
-    rate_limiter: Optional[BaseRateLimiter] = Field(default=None, exclude=True)
+    rate_limiter: BaseRateLimiter | None = Field(default=None, exclude=True)
     "An optional rate limiter to use for limiting the number of requests."
 
-    disable_streaming: Union[bool, Literal["tool_calling"]] = False
+    disable_streaming: bool | Literal["tool_calling"] = False
     """Whether to disable streaming for this model.
 
-    If streaming is bypassed, then ``stream()``/``astream()``/``astream_events()`` will
-    defer to ``invoke()``/``ainvoke()``.
+    If streaming is bypassed, then `stream`/`astream`/`astream_events` will
+    defer to `invoke`/`ainvoke`.
 
-    - If True, will always bypass streaming case.
-    - If ``'tool_calling'``, will bypass streaming case only when the model is called
-      with a ``tools`` keyword argument. In other words, LangChain will automatically
-      switch to non-streaming behavior (``invoke()``) only when the tools argument is
-      provided. This offers the best of both worlds.
-    - If False (default), will always use streaming case if available.
+    - If `True`, will always bypass streaming case.
+    - If `'tool_calling'`, will bypass streaming case only when the model is called
+        with a `tools` keyword argument. In other words, LangChain will automatically
+        switch to non-streaming behavior (`invoke`) only when the tools argument is
+        provided. This offers the best of both worlds.
+    - If `False` (Default), will always use streaming case if available.
 
-    The main reason for this flag is that code might be written using ``.stream()`` and
+    The main reason for this flag is that code might be written using `stream` and
     a user may want to swap out a given model for another model whose the implementation
     does not properly support streaming.
     """
 
-    @model_validator(mode="before")
-    @classmethod
-    def raise_deprecation(cls, values: dict) -> Any:
-        """Raise deprecation warning if callback_manager is used.
+    output_version: str | None = Field(
+        default_factory=from_env("LC_OUTPUT_VERSION", default=None)
+    )
+    """Version of `AIMessage` output format to store in message content.
 
-        Args:
-            values (Dict): Values to validate.
+    `AIMessage.content_blocks` will lazily parse the contents of `content` into a
+    standard format. This flag can be used to additionally store the standard format
+    in message content, e.g., for serialization purposes.
 
-        Returns:
-            Dict: Validated values.
+    Supported values:
 
-        Raises:
-            DeprecationWarning: If callback_manager is used.
-        """
-        if values.get("callback_manager") is not None:
-            warnings.warn(
-                "callback_manager is deprecated. Please use callbacks instead.",
-                DeprecationWarning,
-                stacklevel=5,
-            )
-            values["callbacks"] = values.pop("callback_manager", None)
-        return values
+    - `'v0'`: provider-specific format in content (can lazily-parse with
+        `content_blocks`)
+    - `'v1'`: standardized format in content (consistent with `content_blocks`)
+
+    Partner packages (e.g.,
+    [`langchain-openai`](https://pypi.org/project/langchain-openai)) can also use this
+    field to roll out new content formats in a backward-compatible way.
+
+    !!! version-added "Added in version 1.0"
+
+    """
 
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
@@ -352,7 +346,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     @property
     @override
     def OutputType(self) -> Any:
-        """Get the output type for this runnable."""
+        """Get the output type for this `Runnable`."""
         return AnyMessage
 
     def _convert_input(self, model_input: LanguageModelInput) -> PromptValue:
@@ -372,35 +366,38 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def invoke(
         self,
         input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
+        config: RunnableConfig | None = None,
         *,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
-    ) -> BaseMessage:
+    ) -> AIMessage:
         config = ensure_config(config)
         return cast(
-            "ChatGeneration",
-            self.generate_prompt(
-                [self._convert_input(input)],
-                stop=stop,
-                callbacks=config.get("callbacks"),
-                tags=config.get("tags"),
-                metadata=config.get("metadata"),
-                run_name=config.get("run_name"),
-                run_id=config.pop("run_id", None),
-                **kwargs,
-            ).generations[0][0],
-        ).message
+            "AIMessage",
+            cast(
+                "ChatGeneration",
+                self.generate_prompt(
+                    [self._convert_input(input)],
+                    stop=stop,
+                    callbacks=config.get("callbacks"),
+                    tags=config.get("tags"),
+                    metadata=config.get("metadata"),
+                    run_name=config.get("run_name"),
+                    run_id=config.pop("run_id", None),
+                    **kwargs,
+                ).generations[0][0],
+            ).message,
+        )
 
     @override
     async def ainvoke(
         self,
         input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
+        config: RunnableConfig | None = None,
         *,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
-    ) -> BaseMessage:
+    ) -> AIMessage:
         config = ensure_config(config)
         llm_result = await self.agenerate_prompt(
             [self._convert_input(input)],
@@ -412,15 +409,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             run_id=config.pop("run_id", None),
             **kwargs,
         )
-        return cast("ChatGeneration", llm_result.generations[0][0]).message
+        return cast(
+            "AIMessage", cast("ChatGeneration", llm_result.generations[0][0]).message
+        )
 
     def _should_stream(
         self,
         *,
         async_api: bool,
-        run_manager: Optional[
-            Union[CallbackManagerForLLMRun, AsyncCallbackManagerForLLMRun]
-        ] = None,
+        run_manager: CallbackManagerForLLMRun
+        | AsyncCallbackManagerForLLMRun
+        | None = None,
         **kwargs: Any,
     ) -> bool:
         """Determine if a given model call should hit the streaming API."""
@@ -445,6 +444,11 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         if "stream" in kwargs:
             return kwargs["stream"]
 
+        if "streaming" in self.model_fields_set:
+            streaming_value = getattr(self, "streaming", None)
+            if isinstance(streaming_value, bool):
+                return streaming_value
+
         # Check if any streaming callback handlers have been passed in.
         handlers = run_manager.handlers if run_manager else []
         return any(isinstance(h, _StreamingCallbackHandler) for h in handlers)
@@ -453,15 +457,15 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def stream(
         self,
         input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
+        config: RunnableConfig | None = None,
         *,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
-    ) -> Iterator[BaseMessageChunk]:
+    ) -> Iterator[AIMessageChunk]:
         if not self._should_stream(async_api=False, **{**kwargs, "stream": True}):
-            # model doesn't implement streaming, so use default implementation
+            # Model doesn't implement streaming, so use default implementation
             yield cast(
-                "BaseMessageChunk",
+                "AIMessageChunk",
                 self.invoke(input, config=config, stop=stop, **kwargs),
             )
         else:
@@ -506,16 +510,51 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
             try:
                 input_messages = _normalize_messages(messages)
-                run_id = "-".join((_LC_ID_PREFIX, str(run_manager.run_id)))
+                run_id = "-".join((LC_ID_PREFIX, str(run_manager.run_id)))
+                yielded = False
+                index = -1
+                index_type = ""
                 for chunk in self._stream(input_messages, stop=stop, **kwargs):
                     if chunk.message.id is None:
                         chunk.message.id = run_id
                     chunk.message.response_metadata = _gen_info_and_msg_metadata(chunk)
+                    if self.output_version == "v1":
+                        # Overwrite .content with .content_blocks
+                        chunk.message = _update_message_content_to_blocks(
+                            chunk.message, "v1"
+                        )
+                        for block in cast(
+                            "list[types.ContentBlock]", chunk.message.content
+                        ):
+                            if block["type"] != index_type:
+                                index_type = block["type"]
+                                index = index + 1
+                            if "index" not in block:
+                                block["index"] = index
                     run_manager.on_llm_new_token(
                         cast("str", chunk.message.content), chunk=chunk
                     )
                     chunks.append(chunk)
-                    yield chunk.message
+                    yield cast("AIMessageChunk", chunk.message)
+                    yielded = True
+
+                # Yield a final empty chunk with chunk_position="last" if not yet
+                # yielded
+                if (
+                    yielded
+                    and isinstance(chunk.message, AIMessageChunk)
+                    and not chunk.message.chunk_position
+                ):
+                    empty_content: str | list = (
+                        "" if isinstance(chunk.message.content, str) else []
+                    )
+                    msg_chunk = AIMessageChunk(
+                        content=empty_content, chunk_position="last", id=run_id
+                    )
+                    run_manager.on_llm_new_token(
+                        "", chunk=ChatGenerationChunk(message=msg_chunk)
+                    )
+                    yield msg_chunk
             except BaseException as e:
                 generations_with_error_metadata = _generate_response_from_error(e)
                 chat_generation_chunk = merge_chat_generation_chunks(chunks)
@@ -528,7 +567,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     generations = [generations_with_error_metadata]
                 run_manager.on_llm_error(
                     e,
-                    response=LLMResult(generations=generations),  # type: ignore[arg-type]
+                    response=LLMResult(generations=generations),
                 )
                 raise
 
@@ -544,15 +583,15 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     async def astream(
         self,
         input: LanguageModelInput,
-        config: Optional[RunnableConfig] = None,
+        config: RunnableConfig | None = None,
         *,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
-    ) -> AsyncIterator[BaseMessageChunk]:
+    ) -> AsyncIterator[AIMessageChunk]:
         if not self._should_stream(async_api=True, **{**kwargs, "stream": True}):
             # No async or sync stream is implemented, so fall back to ainvoke
             yield cast(
-                "BaseMessageChunk",
+                "AIMessageChunk",
                 await self.ainvoke(input, config=config, stop=stop, **kwargs),
             )
             return
@@ -599,7 +638,10 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
         try:
             input_messages = _normalize_messages(messages)
-            run_id = "-".join((_LC_ID_PREFIX, str(run_manager.run_id)))
+            run_id = "-".join((LC_ID_PREFIX, str(run_manager.run_id)))
+            yielded = False
+            index = -1
+            index_type = ""
             async for chunk in self._astream(
                 input_messages,
                 stop=stop,
@@ -608,11 +650,42 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 if chunk.message.id is None:
                     chunk.message.id = run_id
                 chunk.message.response_metadata = _gen_info_and_msg_metadata(chunk)
+                if self.output_version == "v1":
+                    # Overwrite .content with .content_blocks
+                    chunk.message = _update_message_content_to_blocks(
+                        chunk.message, "v1"
+                    )
+                    for block in cast(
+                        "list[types.ContentBlock]", chunk.message.content
+                    ):
+                        if block["type"] != index_type:
+                            index_type = block["type"]
+                            index = index + 1
+                        if "index" not in block:
+                            block["index"] = index
                 await run_manager.on_llm_new_token(
                     cast("str", chunk.message.content), chunk=chunk
                 )
                 chunks.append(chunk)
-                yield chunk.message
+                yield cast("AIMessageChunk", chunk.message)
+                yielded = True
+
+            # Yield a final empty chunk with chunk_position="last" if not yet yielded
+            if (
+                yielded
+                and isinstance(chunk.message, AIMessageChunk)
+                and not chunk.message.chunk_position
+            ):
+                empty_content: str | list = (
+                    "" if isinstance(chunk.message.content, str) else []
+                )
+                msg_chunk = AIMessageChunk(
+                    content=empty_content, chunk_position="last", id=run_id
+                )
+                await run_manager.on_llm_new_token(
+                    "", chunk=ChatGenerationChunk(message=msg_chunk)
+                )
+                yield msg_chunk
         except BaseException as e:
             generations_with_error_metadata = _generate_response_from_error(e)
             chat_generation_chunk = merge_chat_generation_chunks(chunks)
@@ -622,7 +695,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 generations = [generations_with_error_metadata]
             await run_manager.on_llm_error(
                 e,
-                response=LLMResult(generations=generations),  # type: ignore[arg-type]
+                response=LLMResult(generations=generations),
             )
             raise
 
@@ -638,7 +711,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
     # --- Custom methods ---
 
-    def _combine_llm_outputs(self, llm_outputs: list[Optional[dict]]) -> dict:  # noqa: ARG002
+    def _combine_llm_outputs(self, llm_outputs: list[dict | None]) -> dict:  # noqa: ARG002
         return {}
 
     def _convert_cached_generations(self, cache_val: list) -> list[ChatGeneration]:
@@ -653,6 +726,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
         Returns:
             List of ChatGeneration objects.
+
         """
         converted_generations = []
         for gen in cache_val:
@@ -666,12 +740,22 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 converted_generations.append(chat_gen)
             else:
                 # Already a ChatGeneration or other expected type
+                if hasattr(gen, "message") and isinstance(gen.message, AIMessage):
+                    # We zero out cost on cache hits
+                    gen.message = gen.message.model_copy(
+                        update={
+                            "usage_metadata": {
+                                **(gen.message.usage_metadata or {}),
+                                "total_cost": 0,
+                            }
+                        }
+                    )
                 converted_generations.append(gen)
         return converted_generations
 
     def _get_invocation_params(
         self,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
     ) -> dict:
         params = self.dict()
@@ -680,7 +764,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
     def _get_ls_params(
         self,
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         **kwargs: Any,
     ) -> LangSmithParams:
         """Get standard params for tracing."""
@@ -697,7 +781,9 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             ls_params["ls_stop"] = stop
 
         # model
-        if hasattr(self, "model") and isinstance(self.model, str):
+        if "model" in kwargs and isinstance(kwargs["model"], str):
+            ls_params["ls_model_name"] = kwargs["model"]
+        elif hasattr(self, "model") and isinstance(self.model, str):
             ls_params["ls_model_name"] = self.model
         elif hasattr(self, "model_name") and isinstance(self.model_name, str):
             ls_params["ls_model_name"] = self.model_name
@@ -716,7 +802,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
         return ls_params
 
-    def _get_llm_string(self, stop: Optional[list[str]] = None, **kwargs: Any) -> str:
+    def _get_llm_string(self, stop: list[str] | None = None, **kwargs: Any) -> str:
         if self.is_lc_serializable():
             params = {**kwargs, "stop": stop}
             param_string = str(sorted(params.items()))
@@ -733,13 +819,13 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def generate(
         self,
         messages: list[list[BaseMessage]],
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         callbacks: Callbacks = None,
         *,
-        tags: Optional[list[str]] = None,
-        metadata: Optional[dict[str, Any]] = None,
-        run_name: Optional[str] = None,
-        run_id: Optional[uuid.UUID] = None,
+        tags: list[str] | None = None,
+        metadata: dict[str, Any] | None = None,
+        run_name: str | None = None,
+        run_id: uuid.UUID | None = None,
         **kwargs: Any,
     ) -> LLMResult:
         """Pass a sequence of prompts to the model and return model generations.
@@ -748,16 +834,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         API.
 
         Use this method when you want to:
-            1. take advantage of batched calls,
-            2. need more output from the model than just the top generated value,
-            3. are building chains that are agnostic to the underlying language model
-                type (e.g., pure text completion models vs chat models).
+
+        1. Take advantage of batched calls,
+        2. Need more output from the model than just the top generated value,
+        3. Are building chains that are agnostic to the underlying language model
+            type (e.g., pure text completion models vs chat models).
 
         Args:
             messages: List of list of messages.
             stop: Stop words to use when generating. Model output is cut off at the
                 first occurrence of any of these substrings.
-            callbacks: Callbacks to pass through. Used for executing additional
+            callbacks: `Callbacks` to pass through. Used for executing additional
                 functionality, such as logging or streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
@@ -767,8 +854,9 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 to the model provider API call.
 
         Returns:
-            An LLMResult, which contains a list of candidate Generations for each input
-                prompt and additional model provider-specific output.
+            An `LLMResult`, which contains a list of candidate `Generations` for each
+                input prompt and additional model provider-specific output.
+
         """
         ls_structured_output_format = kwargs.pop(
             "ls_structured_output_format", None
@@ -825,20 +913,22 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     run_managers[i].on_llm_error(
                         e,
                         response=LLMResult(
-                            generations=[generations_with_error_metadata]  # type: ignore[list-item]
+                            generations=[generations_with_error_metadata]
                         ),
                     )
                 raise
         flattened_outputs = [
-            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[list-item]
+            LLMResult(generations=[res.generations], llm_output=res.llm_output)
             for res in results
         ]
         llm_output = self._combine_llm_outputs([res.llm_output for res in results])
         generations = [res.generations for res in results]
-        output = LLMResult(generations=generations, llm_output=llm_output)  # type: ignore[arg-type]
+        output = LLMResult(generations=generations, llm_output=llm_output)
         if run_managers:
             run_infos = []
-            for manager, flattened_output in zip(run_managers, flattened_outputs):
+            for manager, flattened_output in zip(
+                run_managers, flattened_outputs, strict=False
+            ):
                 manager.on_llm_end(flattened_output)
                 run_infos.append(RunInfo(run_id=manager.run_id))
             output.run = run_infos
@@ -847,13 +937,13 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     async def agenerate(
         self,
         messages: list[list[BaseMessage]],
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         callbacks: Callbacks = None,
         *,
-        tags: Optional[list[str]] = None,
-        metadata: Optional[dict[str, Any]] = None,
-        run_name: Optional[str] = None,
-        run_id: Optional[uuid.UUID] = None,
+        tags: list[str] | None = None,
+        metadata: dict[str, Any] | None = None,
+        run_name: str | None = None,
+        run_id: uuid.UUID | None = None,
         **kwargs: Any,
     ) -> LLMResult:
         """Asynchronously pass a sequence of prompts to a model and return generations.
@@ -862,16 +952,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         API.
 
         Use this method when you want to:
-            1. take advantage of batched calls,
-            2. need more output from the model than just the top generated value,
-            3. are building chains that are agnostic to the underlying language model
-                type (e.g., pure text completion models vs chat models).
+
+        1. Take advantage of batched calls,
+        2. Need more output from the model than just the top generated value,
+        3. Are building chains that are agnostic to the underlying language model
+            type (e.g., pure text completion models vs chat models).
 
         Args:
             messages: List of list of messages.
             stop: Stop words to use when generating. Model output is cut off at the
                 first occurrence of any of these substrings.
-            callbacks: Callbacks to pass through. Used for executing additional
+            callbacks: `Callbacks` to pass through. Used for executing additional
                 functionality, such as logging or streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
@@ -881,8 +972,9 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 to the model provider API call.
 
         Returns:
-            An LLMResult, which contains a list of candidate Generations for each input
-                prompt and additional model provider-specific output.
+            An `LLMResult`, which contains a list of candidate `Generations` for each
+                input prompt and additional model provider-specific output.
+
         """
         ls_structured_output_format = kwargs.pop(
             "ls_structured_output_format", None
@@ -944,7 +1036,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     await run_managers[i].on_llm_error(
                         res,
                         response=LLMResult(
-                            generations=[generations_with_error_metadata]  # type: ignore[list-item]
+                            generations=[generations_with_error_metadata]
                         ),
                     )
                 exceptions.append(res)
@@ -954,27 +1046,27 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     *[
                         run_manager.on_llm_end(
                             LLMResult(
-                                generations=[res.generations],  # type: ignore[list-item, union-attr]
+                                generations=[res.generations],  # type: ignore[union-attr]
                                 llm_output=res.llm_output,  # type: ignore[union-attr]
                             )
                         )
-                        for run_manager, res in zip(run_managers, results)
+                        for run_manager, res in zip(run_managers, results, strict=False)
                         if not isinstance(res, Exception)
                     ]
                 )
             raise exceptions[0]
         flattened_outputs = [
-            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[list-item, union-attr]
+            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[union-attr]
             for res in results
         ]
         llm_output = self._combine_llm_outputs([res.llm_output for res in results])  # type: ignore[union-attr]
         generations = [res.generations for res in results]  # type: ignore[union-attr]
-        output = LLMResult(generations=generations, llm_output=llm_output)  # type: ignore[arg-type]
+        output = LLMResult(generations=generations, llm_output=llm_output)
         await asyncio.gather(
             *[
                 run_manager.on_llm_end(flattened_output)
                 for run_manager, flattened_output in zip(
-                    run_managers, flattened_outputs
+                    run_managers, flattened_outputs, strict=False
                 )
             ]
         )
@@ -988,7 +1080,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def generate_prompt(
         self,
         prompts: list[PromptValue],
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         callbacks: Callbacks = None,
         **kwargs: Any,
     ) -> LLMResult:
@@ -999,7 +1091,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     async def agenerate_prompt(
         self,
         prompts: list[PromptValue],
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         callbacks: Callbacks = None,
         **kwargs: Any,
     ) -> LLMResult:
@@ -1011,8 +1103,8 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def _generate_with_cache(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
         llm_cache = self.cache if isinstance(self.cache, BaseCache) else get_llm_cache()
@@ -1048,15 +1140,53 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             **kwargs,
         ):
             chunks: list[ChatGenerationChunk] = []
+            run_id: str | None = (
+                f"{LC_ID_PREFIX}-{run_manager.run_id}" if run_manager else None
+            )
+            yielded = False
+            index = -1
+            index_type = ""
             for chunk in self._stream(messages, stop=stop, **kwargs):
                 chunk.message.response_metadata = _gen_info_and_msg_metadata(chunk)
+                if self.output_version == "v1":
+                    # Overwrite .content with .content_blocks
+                    chunk.message = _update_message_content_to_blocks(
+                        chunk.message, "v1"
+                    )
+                    for block in cast(
+                        "list[types.ContentBlock]", chunk.message.content
+                    ):
+                        if block["type"] != index_type:
+                            index_type = block["type"]
+                            index = index + 1
+                        if "index" not in block:
+                            block["index"] = index
                 if run_manager:
                     if chunk.message.id is None:
-                        chunk.message.id = f"{_LC_ID_PREFIX}-{run_manager.run_id}"
+                        chunk.message.id = run_id
                     run_manager.on_llm_new_token(
                         cast("str", chunk.message.content), chunk=chunk
                     )
                 chunks.append(chunk)
+                yielded = True
+
+            # Yield a final empty chunk with chunk_position="last" if not yet yielded
+            if (
+                yielded
+                and isinstance(chunk.message, AIMessageChunk)
+                and not chunk.message.chunk_position
+            ):
+                empty_content: str | list = (
+                    "" if isinstance(chunk.message.content, str) else []
+                )
+                chunk = ChatGenerationChunk(
+                    message=AIMessageChunk(
+                        content=empty_content, chunk_position="last", id=run_id
+                    )
+                )
+                if run_manager:
+                    run_manager.on_llm_new_token("", chunk=chunk)
+                chunks.append(chunk)
             result = generate_from_stream(iter(chunks))
         elif inspect.signature(self._generate).parameters.get("run_manager"):
             result = self._generate(
@@ -1065,10 +1195,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         else:
             result = self._generate(messages, stop=stop, **kwargs)
 
+        if self.output_version == "v1":
+            # Overwrite .content with .content_blocks
+            for generation in result.generations:
+                generation.message = _update_message_content_to_blocks(
+                    generation.message, "v1"
+                )
+
         # Add response metadata to each generation
         for idx, generation in enumerate(result.generations):
             if run_manager and generation.message.id is None:
-                generation.message.id = f"{_LC_ID_PREFIX}-{run_manager.run_id}-{idx}"
+                generation.message.id = f"{LC_ID_PREFIX}-{run_manager.run_id}-{idx}"
             generation.message.response_metadata = _gen_info_and_msg_metadata(
                 generation
             )
@@ -1084,8 +1221,8 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     async def _agenerate_with_cache(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: AsyncCallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
         llm_cache = self.cache if isinstance(self.cache, BaseCache) else get_llm_cache()
@@ -1121,15 +1258,53 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             **kwargs,
         ):
             chunks: list[ChatGenerationChunk] = []
+            run_id: str | None = (
+                f"{LC_ID_PREFIX}-{run_manager.run_id}" if run_manager else None
+            )
+            yielded = False
+            index = -1
+            index_type = ""
             async for chunk in self._astream(messages, stop=stop, **kwargs):
                 chunk.message.response_metadata = _gen_info_and_msg_metadata(chunk)
+                if self.output_version == "v1":
+                    # Overwrite .content with .content_blocks
+                    chunk.message = _update_message_content_to_blocks(
+                        chunk.message, "v1"
+                    )
+                    for block in cast(
+                        "list[types.ContentBlock]", chunk.message.content
+                    ):
+                        if block["type"] != index_type:
+                            index_type = block["type"]
+                            index = index + 1
+                        if "index" not in block:
+                            block["index"] = index
                 if run_manager:
                     if chunk.message.id is None:
-                        chunk.message.id = f"{_LC_ID_PREFIX}-{run_manager.run_id}"
+                        chunk.message.id = run_id
                     await run_manager.on_llm_new_token(
                         cast("str", chunk.message.content), chunk=chunk
                     )
                 chunks.append(chunk)
+                yielded = True
+
+            # Yield a final empty chunk with chunk_position="last" if not yet yielded
+            if (
+                yielded
+                and isinstance(chunk.message, AIMessageChunk)
+                and not chunk.message.chunk_position
+            ):
+                empty_content: str | list = (
+                    "" if isinstance(chunk.message.content, str) else []
+                )
+                chunk = ChatGenerationChunk(
+                    message=AIMessageChunk(
+                        content=empty_content, chunk_position="last", id=run_id
+                    )
+                )
+                if run_manager:
+                    await run_manager.on_llm_new_token("", chunk=chunk)
+                chunks.append(chunk)
             result = generate_from_stream(iter(chunks))
         elif inspect.signature(self._agenerate).parameters.get("run_manager"):
             result = await self._agenerate(
@@ -1138,10 +1313,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         else:
             result = await self._agenerate(messages, stop=stop, **kwargs)
 
+        if self.output_version == "v1":
+            # Overwrite .content with .content_blocks
+            for generation in result.generations:
+                generation.message = _update_message_content_to_blocks(
+                    generation.message, "v1"
+                )
+
         # Add response metadata to each generation
         for idx, generation in enumerate(result.generations):
             if run_manager and generation.message.id is None:
-                generation.message.id = f"{_LC_ID_PREFIX}-{run_manager.run_id}-{idx}"
+                generation.message.id = f"{LC_ID_PREFIX}-{run_manager.run_id}-{idx}"
             generation.message.response_metadata = _gen_info_and_msg_metadata(
                 generation
             )
@@ -1158,20 +1340,40 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def _generate(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
-        """Top Level call."""
+        """Generate the result.
+
+        Args:
+            messages: The messages to generate from.
+            stop: Optional list of stop words to use when generating.
+            run_manager: Optional callback manager to use for this call.
+            **kwargs: Additional keyword arguments to pass to the model.
+
+        Returns:
+            The chat result.
+        """
 
     async def _agenerate(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: AsyncCallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
-        """Top Level call."""
+        """Generate the result.
+
+        Args:
+            messages: The messages to generate from.
+            stop: Optional list of stop words to use when generating.
+            run_manager: Optional callback manager to use for this call.
+            **kwargs: Additional keyword arguments to pass to the model.
+
+        Returns:
+            The chat result.
+        """
         return await run_in_executor(
             None,
             self._generate,
@@ -1184,19 +1386,41 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def _stream(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> Iterator[ChatGenerationChunk]:
+        """Stream the output of the model.
+
+        Args:
+            messages: The messages to generate from.
+            stop: Optional list of stop words to use when generating.
+            run_manager: Optional callback manager to use for this call.
+            **kwargs: Additional keyword arguments to pass to the model.
+
+        Yields:
+            The chat generation chunks.
+        """
         raise NotImplementedError
 
     async def _astream(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: AsyncCallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> AsyncIterator[ChatGenerationChunk]:
+        """Stream the output of the model.
+
+        Args:
+            messages: The messages to generate from.
+            stop: Optional list of stop words to use when generating.
+            run_manager: Optional callback manager to use for this call.
+            **kwargs: Additional keyword arguments to pass to the model.
+
+        Yields:
+            The chat generation chunks.
+        """
         iterator = await run_in_executor(
             None,
             self._stream,
@@ -1217,40 +1441,10 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 break
             yield item  # type: ignore[misc]
 
-    @deprecated("0.1.7", alternative="invoke", removal="1.0")
-    def __call__(
-        self,
-        messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        callbacks: Callbacks = None,
-        **kwargs: Any,
-    ) -> BaseMessage:
-        """Call the model.
-
-        Args:
-            messages: List of messages.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: Callbacks to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
-
-        Returns:
-            The model output message.
-        """
-        generation = self.generate(
-            [messages], stop=stop, callbacks=callbacks, **kwargs
-        ).generations[0][0]
-        if isinstance(generation, ChatGeneration):
-            return generation.message
-        msg = "Unexpected generation type"
-        raise ValueError(msg)
-
     async def _call_async(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
+        stop: list[str] | None = None,
         callbacks: Callbacks = None,
         **kwargs: Any,
     ) -> BaseMessage:
@@ -1263,86 +1457,6 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
         msg = "Unexpected generation type"
         raise ValueError(msg)
 
-    @deprecated("0.1.7", alternative="invoke", removal="1.0")
-    def call_as_llm(
-        self, message: str, stop: Optional[list[str]] = None, **kwargs: Any
-    ) -> str:
-        """Call the model.
-
-        Args:
-            message: The input message.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
-
-        Returns:
-            The model output string.
-        """
-        return self.predict(message, stop=stop, **kwargs)
-
-    @deprecated("0.1.7", alternative="invoke", removal="1.0")
-    @override
-    def predict(
-        self, text: str, *, stop: Optional[Sequence[str]] = None, **kwargs: Any
-    ) -> str:
-        """Predict the next message.
-
-        Args:
-            text: The input message.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
-
-        Returns:
-            The predicted output string.
-        """
-        stop_ = None if stop is None else list(stop)
-        result = self([HumanMessage(content=text)], stop=stop_, **kwargs)
-        if isinstance(result.content, str):
-            return result.content
-        msg = "Cannot use predict when output is not a string."
-        raise ValueError(msg)
-
-    @deprecated("0.1.7", alternative="invoke", removal="1.0")
-    @override
-    def predict_messages(
-        self,
-        messages: list[BaseMessage],
-        *,
-        stop: Optional[Sequence[str]] = None,
-        **kwargs: Any,
-    ) -> BaseMessage:
-        stop_ = None if stop is None else list(stop)
-        return self(messages, stop=stop_, **kwargs)
-
-    @deprecated("0.1.7", alternative="ainvoke", removal="1.0")
-    @override
-    async def apredict(
-        self, text: str, *, stop: Optional[Sequence[str]] = None, **kwargs: Any
-    ) -> str:
-        stop_ = None if stop is None else list(stop)
-        result = await self._call_async(
-            [HumanMessage(content=text)], stop=stop_, **kwargs
-        )
-        if isinstance(result.content, str):
-            return result.content
-        msg = "Cannot use predict when output is not a string."
-        raise ValueError(msg)
-
-    @deprecated("0.1.7", alternative="ainvoke", removal="1.0")
-    @override
-    async def apredict_messages(
-        self,
-        messages: list[BaseMessage],
-        *,
-        stop: Optional[Sequence[str]] = None,
-        **kwargs: Any,
-    ) -> BaseMessage:
-        stop_ = None if stop is None else list(stop)
-        return await self._call_async(messages, stop=stop_, **kwargs)
-
     @property
     @abstractmethod
     def _llm_type(self) -> str:
@@ -1358,12 +1472,12 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
     def bind_tools(
         self,
         tools: Sequence[
-            Union[typing.Dict[str, Any], type, Callable, BaseTool]  # noqa: UP006
+            typing.Dict[str, Any] | type | Callable | BaseTool  # noqa: UP006
         ],
         *,
-        tool_choice: Optional[Union[str]] = None,
+        tool_choice: str | None = None,
         **kwargs: Any,
-    ) -> Runnable[LanguageModelInput, BaseMessage]:
+    ) -> Runnable[LanguageModelInput, AIMessage]:
         """Bind tools to the model.
 
         Args:
@@ -1372,16 +1486,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
         Returns:
             A Runnable that returns a message.
+
         """
         raise NotImplementedError
 
     def with_structured_output(
         self,
-        schema: Union[typing.Dict, type],  # noqa: UP006
+        schema: typing.Dict | type,  # noqa: UP006
         *,
         include_raw: bool = False,
         **kwargs: Any,
-    ) -> Runnable[LanguageModelInput, Union[typing.Dict, BaseModel]]:  # noqa: UP006
+    ) -> Runnable[LanguageModelInput, typing.Dict | BaseModel]:  # noqa: UP006
         """Model wrapper that returns outputs formatted to match the given schema.
 
         Args:
@@ -1389,102 +1504,130 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 
                 - an OpenAI function/tool schema,
                 - a JSON Schema,
-                - a TypedDict class,
+                - a `TypedDict` class,
                 - or a Pydantic class.
 
-                If ``schema`` is a Pydantic class then the model output will be a
+                If `schema` is a Pydantic class then the model output will be a
                 Pydantic instance of that class, and the model-generated fields will be
                 validated by the Pydantic class. Otherwise the model output will be a
-                dict and will not be validated. See :meth:`langchain_core.utils.function_calling.convert_to_openai_tool`
-                for more on how to properly specify types and descriptions of
-                schema fields when specifying a Pydantic or TypedDict class.
+                dict and will not be validated.
+
+                See `langchain_core.utils.function_calling.convert_to_openai_tool` for
+                more on how to properly specify types and descriptions of schema fields
+                when specifying a Pydantic or `TypedDict` class.
 
             include_raw:
-                If False then only the parsed structured output is returned. If
-                an error occurs during model output parsing it will be raised. If True
-                then both the raw model response (a BaseMessage) and the parsed model
+                If `False` then only the parsed structured output is returned. If
+                an error occurs during model output parsing it will be raised. If `True`
+                then both the raw model response (a `BaseMessage`) and the parsed model
                 response will be returned. If an error occurs during output parsing it
-                will be caught and returned as well. The final output is always a dict
-                with keys ``'raw'``, ``'parsed'``, and ``'parsing_error'``.
+                will be caught and returned as well.
+
+                The final output is always a `dict` with keys `'raw'`, `'parsed'`, and
+                `'parsing_error'`.
+
+        Raises:
+            ValueError: If there are any unsupported `kwargs`.
+            NotImplementedError: If the model does not implement
+                `with_structured_output()`.
 
         Returns:
-            A Runnable that takes same inputs as a :class:`langchain_core.language_models.chat.BaseChatModel`.
+            A `Runnable` that takes same inputs as a
+                `langchain_core.language_models.chat.BaseChatModel`. If `include_raw` is
+                `False` and `schema` is a Pydantic class, `Runnable` outputs an instance
+                of `schema` (i.e., a Pydantic object). Otherwise, if `include_raw` is
+                `False` then `Runnable` outputs a `dict`.
+
+                If `include_raw` is `True`, then `Runnable` outputs a `dict` with keys:
 
-            If ``include_raw`` is False and ``schema`` is a Pydantic class, Runnable outputs
-            an instance of ``schema`` (i.e., a Pydantic object).
+                - `'raw'`: `BaseMessage`
+                - `'parsed'`: `None` if there was a parsing error, otherwise the type
+                    depends on the `schema` as described above.
+                - `'parsing_error'`: `BaseException | None`
 
-            Otherwise, if ``include_raw`` is False then Runnable outputs a dict.
+        Example: Pydantic schema (`include_raw=False`):
 
-            If ``include_raw`` is True, then Runnable outputs a dict with keys:
+        ```python
+        from pydantic import BaseModel
 
-            - ``'raw'``: BaseMessage
-            - ``'parsed'``: None if there was a parsing error, otherwise the type depends on the ``schema`` as described above.
-            - ``'parsing_error'``: Optional[BaseException]
 
-        Example: Pydantic schema (include_raw=False):
-            .. code-block:: python
+        class AnswerWithJustification(BaseModel):
+            '''An answer to the user question along with justification for the answer.'''
 
-                from pydantic import BaseModel
+            answer: str
+            justification: str
 
-                class AnswerWithJustification(BaseModel):
-                    '''An answer to the user question along with justification for the answer.'''
-                    answer: str
-                    justification: str
 
-                llm = ChatModel(model="model-name", temperature=0)
-                structured_llm = llm.with_structured_output(AnswerWithJustification)
+        model = ChatModel(model="model-name", temperature=0)
+        structured_model = model.with_structured_output(AnswerWithJustification)
 
-                structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers")
+        structured_model.invoke(
+            "What weighs more a pound of bricks or a pound of feathers"
+        )
+
+        # -> AnswerWithJustification(
+        #     answer='They weigh the same',
+        #     justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.'
+        # )
+        ```
+
+        Example: Pydantic schema (`include_raw=True`):
+
+        ```python
+        from pydantic import BaseModel
+
+
+        class AnswerWithJustification(BaseModel):
+            '''An answer to the user question along with justification for the answer.'''
 
-                # -> AnswerWithJustification(
-                #     answer='They weigh the same',
-                #     justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.'
-                # )
+            answer: str
+            justification: str
 
-        Example: Pydantic schema (include_raw=True):
-            .. code-block:: python
 
-                from pydantic import BaseModel
+        model = ChatModel(model="model-name", temperature=0)
+        structured_model = model.with_structured_output(
+            AnswerWithJustification, include_raw=True
+        )
 
-                class AnswerWithJustification(BaseModel):
-                    '''An answer to the user question along with justification for the answer.'''
-                    answer: str
-                    justification: str
+        structured_model.invoke(
+            "What weighs more a pound of bricks or a pound of feathers"
+        )
+        # -> {
+        #     'raw': AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_Ao02pnFYXD6GN1yzc0uXPsvF', 'function': {'arguments': '{"answer":"They weigh the same.","justification":"Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ."}', 'name': 'AnswerWithJustification'}, 'type': 'function'}]}),
+        #     'parsed': AnswerWithJustification(answer='They weigh the same.', justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.'),
+        #     'parsing_error': None
+        # }
+        ```
 
-                llm = ChatModel(model="model-name", temperature=0)
-                structured_llm = llm.with_structured_output(AnswerWithJustification, include_raw=True)
+        Example: `dict` schema (`include_raw=False`):
 
-                structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers")
-                # -> {
-                #     'raw': AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_Ao02pnFYXD6GN1yzc0uXPsvF', 'function': {'arguments': '{"answer":"They weigh the same.","justification":"Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ."}', 'name': 'AnswerWithJustification'}, 'type': 'function'}]}),
-                #     'parsed': AnswerWithJustification(answer='They weigh the same.', justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.'),
-                #     'parsing_error': None
-                # }
+        ```python
+        from pydantic import BaseModel
+        from langchain_core.utils.function_calling import convert_to_openai_tool
 
-        Example: Dict schema (include_raw=False):
-            .. code-block:: python
 
-                from pydantic import BaseModel
-                from langchain_core.utils.function_calling import convert_to_openai_tool
+        class AnswerWithJustification(BaseModel):
+            '''An answer to the user question along with justification for the answer.'''
 
-                class AnswerWithJustification(BaseModel):
-                    '''An answer to the user question along with justification for the answer.'''
-                    answer: str
-                    justification: str
+            answer: str
+            justification: str
 
-                dict_schema = convert_to_openai_tool(AnswerWithJustification)
-                llm = ChatModel(model="model-name", temperature=0)
-                structured_llm = llm.with_structured_output(dict_schema)
 
-                structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers")
-                # -> {
-                #     'answer': 'They weigh the same',
-                #     'justification': 'Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume and density of the two substances differ.'
-                # }
+        dict_schema = convert_to_openai_tool(AnswerWithJustification)
+        model = ChatModel(model="model-name", temperature=0)
+        structured_model = model.with_structured_output(dict_schema)
 
-        .. versionchanged:: 0.2.26
+        structured_model.invoke(
+            "What weighs more a pound of bricks or a pound of feathers"
+        )
+        # -> {
+        #     'answer': 'They weigh the same',
+        #     'justification': 'Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume and density of the two substances differ.'
+        # }
+        ```
 
-                Added support for TypedDict class.
+        !!! warning "Behavior changed in 0.2.26"
+            Added support for TypedDict class.
 
         """  # noqa: E501
         _ = kwargs.pop("method", None)
@@ -1493,11 +1636,6 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
             msg = f"Received unsupported arguments {kwargs}"
             raise ValueError(msg)
 
-        from langchain_core.output_parsers.openai_tools import (
-            JsonOutputKeyToolsParser,
-            PydanticToolsParser,
-        )
-
         if type(self).bind_tools is BaseChatModel.bind_tools:
             msg = "with_structured_output is not implemented for this model."
             raise NotImplementedError(msg)
@@ -1534,15 +1672,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
 class SimpleChatModel(BaseChatModel):
     """Simplified implementation for a chat model to inherit from.
 
-    **Note** This implementation is primarily here for backwards compatibility.
-        For new implementations, please use `BaseChatModel` directly.
+    !!! note
+        This implementation is primarily here for backwards compatibility. For new
+        implementations, please use `BaseChatModel` directly.
+
     """
 
     def _generate(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
         output_str = self._call(messages, stop=stop, run_manager=run_manager, **kwargs)
@@ -1554,8 +1694,8 @@ class SimpleChatModel(BaseChatModel):
     def _call(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[CallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: CallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> str:
         """Simpler interface."""
@@ -1563,8 +1703,8 @@ class SimpleChatModel(BaseChatModel):
     async def _agenerate(
         self,
         messages: list[BaseMessage],
-        stop: Optional[list[str]] = None,
-        run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
+        stop: list[str] | None = None,
+        run_manager: AsyncCallbackManagerForLLMRun | None = None,
         **kwargs: Any,
     ) -> ChatResult:
         return await run_in_executor(
@@ -1578,7 +1718,7 @@ class SimpleChatModel(BaseChatModel):
 
 
 def _gen_info_and_msg_metadata(
-    generation: Union[ChatGeneration, ChatGenerationChunk],
+    generation: ChatGeneration | ChatGenerationChunk,
 ) -> dict:
     return {
         **(generation.generation_info or {}),