langchain-ollama 0.2.0.dev1__py3-none-any.whl → 0.2.2__py3-none-any.whl

langchain_ollama/__init__.py

@@ -1,3 +1,9 @@
+"""This is the langchain_ollama package.
+
+It provides infrastructure for interacting with the Ollama service.
+"""
+
+
 from importlib import metadata
 
 from langchain_ollama.chat_models import ChatOllama

langchain_ollama/chat_models.py

@@ -1,5 +1,7 @@
 """Ollama chat models."""
 
+import json
+from operator import itemgetter
 from typing import (
     Any,
     AsyncIterator,
@@ -21,6 +23,7 @@ from langchain_core.callbacks import (
     CallbackManagerForLLMRun,
 )
 from langchain_core.callbacks.manager import AsyncCallbackManagerForLLMRun
+from langchain_core.exceptions import OutputParserException
 from langchain_core.language_models import LanguageModelInput
 from langchain_core.language_models.chat_models import BaseChatModel, LangSmithParams
 from langchain_core.messages import (
@@ -34,13 +37,24 @@ from langchain_core.messages import (
 )
 from langchain_core.messages.ai import UsageMetadata
 from langchain_core.messages.tool import tool_call
+from langchain_core.output_parsers import (
+    JsonOutputKeyToolsParser,
+    JsonOutputParser,
+    PydanticOutputParser,
+    PydanticToolsParser,
+)
 from langchain_core.outputs import ChatGeneration, ChatGenerationChunk, ChatResult
-from langchain_core.runnables import Runnable
+from langchain_core.runnables import Runnable, RunnableMap, RunnablePassthrough
 from langchain_core.tools import BaseTool
+from langchain_core.utils.function_calling import (
+    _convert_any_typed_dicts_to_pydantic as convert_any_typed_dicts_to_pydantic,
+)
 from langchain_core.utils.function_calling import convert_to_openai_tool
+from langchain_core.utils.pydantic import TypeBaseModel, is_basemodel_subclass
 from ollama import AsyncClient, Client, Message, Options
-from pydantic import PrivateAttr, model_validator
-from typing_extensions import Self
+from pydantic import BaseModel, PrivateAttr, model_validator
+from pydantic.json_schema import JsonSchemaValue
+from typing_extensions import Self, is_typeddict
 
 
 def _get_usage_metadata_from_generation_info(
@@ -60,19 +74,85 @@ def _get_usage_metadata_from_generation_info(
     return None
 
 
+def _parse_json_string(
+    json_string: str, raw_tool_call: dict[str, Any], skip: bool
+) -> Any:
+    """Attempt to parse a JSON string for tool calling.
+
+    Args:
+        json_string: JSON string to parse.
+        skip: Whether to ignore parsing errors and return the value anyways.
+        raw_tool_call: Raw tool call to include in error message.
+
+    Returns:
+        The parsed JSON string.
+
+    Raises:
+        OutputParserException: If the JSON string wrong invalid and skip=False.
+    """
+    try:
+        return json.loads(json_string)
+    except json.JSONDecodeError as e:
+        if skip:
+            return json_string
+        msg = (
+            f"Function {raw_tool_call['function']['name']} arguments:\n\n"
+            f"{raw_tool_call['function']['arguments']}\n\nare not valid JSON. "
+            f"Received JSONDecodeError {e}"
+        )
+        raise OutputParserException(msg) from e
+    except TypeError as e:
+        if skip:
+            return json_string
+        msg = (
+            f"Function {raw_tool_call['function']['name']} arguments:\n\n"
+            f"{raw_tool_call['function']['arguments']}\n\nare not a string or a "
+            f"dictionary. Received TypeError {e}"
+        )
+        raise OutputParserException(msg) from e
+
+
+def _parse_arguments_from_tool_call(
+    raw_tool_call: dict[str, Any],
+) -> Optional[dict[str, Any]]:
+    """Parse arguments by trying to parse any shallowly nested string-encoded JSON.
+
+    Band-aid fix for issue in Ollama with inconsistent tool call argument structure.
+    Should be removed/changed if fixed upstream.
+    See https://github.com/ollama/ollama/issues/6155
+    """
+    if "function" not in raw_tool_call:
+        return None
+    arguments = raw_tool_call["function"]["arguments"]
+    parsed_arguments = {}
+    if isinstance(arguments, dict):
+        for key, value in arguments.items():
+            if isinstance(value, str):
+                parsed_arguments[key] = _parse_json_string(
+                    value, skip=True, raw_tool_call=raw_tool_call
+                )
+            else:
+                parsed_arguments[key] = value
+    else:
+        parsed_arguments = _parse_json_string(
+            arguments, skip=False, raw_tool_call=raw_tool_call
+        )
+    return parsed_arguments
+
+
 def _get_tool_calls_from_response(
     response: Mapping[str, Any],
 ) -> List[ToolCall]:
     """Get tool calls from ollama response."""
     tool_calls = []
     if "message" in response:
-        if "tool_calls" in response["message"]:
-            for tc in response["message"]["tool_calls"]:
+        if raw_tool_calls := response["message"].get("tool_calls"):
+            for tc in raw_tool_calls:
                 tool_calls.append(
                     tool_call(
                         id=str(uuid4()),
                         name=tc["function"]["name"],
-                        args=tc["function"]["arguments"],
+                        args=_parse_arguments_from_tool_call(tc) or {},
                     )
                 )
     return tool_calls
@@ -89,10 +169,16 @@ def _lc_tool_call_to_openai_tool_call(tool_call: ToolCall) -> dict:
     }
 
 
+def _is_pydantic_class(obj: Any) -> bool:
+    return isinstance(obj, type) and is_basemodel_subclass(obj)
+
+
 class ChatOllama(BaseChatModel):
-    """Ollama chat model integration.
+    r"""Ollama chat model integration.
+
+    .. dropdown:: Setup
+        :open:
 
-    Setup:
         Install ``langchain-ollama`` and download any models you want to use from ollama.
 
         .. code-block:: bash
@@ -220,8 +306,6 @@ class ChatOllama(BaseChatModel):
             '{"location": "Pune, India", "time_of_day": "morning"}'
 
     Tool Calling:
-        .. warning::
-            Ollama currently does not support streaming for tools
 
         .. code-block:: python
 
@@ -315,8 +399,8 @@ class ChatOllama(BaseChatModel):
     to more diverse text, while a lower value (e.g., 0.5) will
     generate more focused and conservative text. (Default: 0.9)"""
 
-    format: Literal["", "json"] = ""
-    """Specify the format of the output (options: json)"""
+    format: Optional[Union[Literal["", "json"], JsonSchemaValue]] = None
+    """Specify the format of the output (options: "json", JSON schema)."""
 
     keep_alive: Optional[Union[int, str]] = None
     """How long the model will stay loaded into memory."""
@@ -325,27 +409,36 @@ class ChatOllama(BaseChatModel):
     """Base url the model is hosted under."""
 
     client_kwargs: Optional[dict] = {}
-    """Additional kwargs to pass to the httpx Client. 
+    """Additional kwargs to pass to the httpx Client.
     For a full list of the params, see [this link](https://pydoc.dev/httpx/latest/httpx.Client.html)
     """
 
-    _client: Client = PrivateAttr(default=None)
+    _client: Client = PrivateAttr(default=None)  # type: ignore
     """
     The client to use for making requests.
     """
 
-    _async_client: AsyncClient = PrivateAttr(default=None)
+    _async_client: AsyncClient = PrivateAttr(default=None)  # type: ignore
     """
     The async client to use for making requests.
     """
 
-    @property
-    def _default_params(self) -> Dict[str, Any]:
-        """Get the default parameters for calling Ollama."""
-        return {
-            "model": self.model,
-            "format": self.format,
-            "options": {
+    def _chat_params(
+        self,
+        messages: List[BaseMessage],
+        stop: Optional[List[str]] = None,
+        **kwargs: Any,
+    ) -> Dict[str, Any]:
+        ollama_messages = self._convert_messages_to_ollama_messages(messages)
+
+        if self.stop is not None and stop is not None:
+            raise ValueError("`stop` found in both the input and default params.")
+        elif self.stop is not None:
+            stop = self.stop
+
+        options_dict = kwargs.pop(
+            "options",
+            {
                 "mirostat": self.mirostat,
                 "mirostat_eta": self.mirostat_eta,
                 "mirostat_tau": self.mirostat_tau,
@@ -357,14 +450,28 @@ class ChatOllama(BaseChatModel):
                 "repeat_penalty": self.repeat_penalty,
                 "temperature": self.temperature,
                 "seed": self.seed,
-                "stop": self.stop,
+                "stop": self.stop if stop is None else stop,
                 "tfs_z": self.tfs_z,
                 "top_k": self.top_k,
                 "top_p": self.top_p,
             },
-            "keep_alive": self.keep_alive,
+        )
+
+        params = {
+            "messages": ollama_messages,
+            "stream": kwargs.pop("stream", True),
+            "model": kwargs.pop("model", self.model),
+            "format": kwargs.pop("format", self.format),
+            "options": Options(**options_dict),
+            "keep_alive": kwargs.pop("keep_alive", self.keep_alive),
+            **kwargs,
         }
 
+        if tools := kwargs.get("tools"):
+            params["tools"] = tools
+
+        return params
+
     @model_validator(mode="after")
     def _set_clients(self) -> Self:
         """Set clients to use for ollama."""
@@ -462,37 +569,13 @@ class ChatOllama(BaseChatModel):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> AsyncIterator[Union[Mapping[str, Any], str]]:
-        ollama_messages = self._convert_messages_to_ollama_messages(messages)
+        chat_params = self._chat_params(messages, stop, **kwargs)
 
-        stop = stop if stop is not None else self.stop
-
-        params = self._default_params
-
-        for key in self._default_params:
-            if key in kwargs:
-                params[key] = kwargs[key]
-
-        params["options"]["stop"] = stop
-        if "tools" in kwargs:
-            yield await self._async_client.chat(
-                model=params["model"],
-                messages=ollama_messages,
-                stream=False,
-                options=Options(**params["options"]),
-                keep_alive=params["keep_alive"],
-                format=params["format"],
-                tools=kwargs["tools"],
-            )  # type:ignore
-        else:
-            async for part in await self._async_client.chat(
-                model=params["model"],
-                messages=ollama_messages,
-                stream=True,
-                options=Options(**params["options"]),
-                keep_alive=params["keep_alive"],
-                format=params["format"],
-            ):  # type:ignore
+        if chat_params["stream"]:
+            async for part in await self._async_client.chat(**chat_params):
                 yield part
+        else:
+            yield await self._async_client.chat(**chat_params)
 
     def _create_chat_stream(
         self,
@@ -500,36 +583,12 @@ class ChatOllama(BaseChatModel):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> Iterator[Union[Mapping[str, Any], str]]:
-        ollama_messages = self._convert_messages_to_ollama_messages(messages)
-
-        stop = stop if stop is not None else self.stop
-
-        params = self._default_params
+        chat_params = self._chat_params(messages, stop, **kwargs)
 
-        for key in self._default_params:
-            if key in kwargs:
-                params[key] = kwargs[key]
-
-        params["options"]["stop"] = stop
-        if "tools" in kwargs:
-            yield self._client.chat(
-                model=params["model"],
-                messages=ollama_messages,
-                stream=False,
-                options=Options(**params["options"]),
-                keep_alive=params["keep_alive"],
-                format=params["format"],
-                tools=kwargs["tools"],
-            )
+        if chat_params["stream"]:
+            yield from self._client.chat(**chat_params)
         else:
-            yield from self._client.chat(
-                model=params["model"],
-                messages=ollama_messages,
-                stream=True,
-                options=Options(**params["options"]),
-                keep_alive=params["keep_alive"],
-                format=params["format"],
-            )
+            yield self._client.chat(**chat_params)
 
     def _chat_stream_with_aggregation(
         self,
@@ -748,6 +807,8 @@ class ChatOllama(BaseChatModel):
     def bind_tools(
         self,
         tools: Sequence[Union[Dict[str, Any], Type, Callable, BaseTool]],
+        *,
+        tool_choice: Optional[Union[dict, str, Literal["auto", "any"], bool]] = None,
         **kwargs: Any,
     ) -> Runnable[LanguageModelInput, BaseMessage]:
         """Bind tool-like objects to this chat model.
@@ -758,8 +819,329 @@ class ChatOllama(BaseChatModel):
             tools: A list of tool definitions to bind to this chat model.
                 Supports any tool definition handled by
                 :meth:`langchain_core.utils.function_calling.convert_to_openai_tool`.
+            tool_choice: If provided, which tool for model to call. **This parameter
+                is currently ignored as it is not supported by Ollama.**
             kwargs: Any additional parameters are passed directly to
                 ``self.bind(**kwargs)``.
         """  # noqa: E501
         formatted_tools = [convert_to_openai_tool(tool) for tool in tools]
         return super().bind(tools=formatted_tools, **kwargs)
+
+    def with_structured_output(
+        self,
+        schema: Union[Dict, type],
+        *,
+        method: Literal[
+            "function_calling", "json_mode", "json_schema"
+        ] = "function_calling",
+        include_raw: bool = False,
+        **kwargs: Any,
+    ) -> Runnable[LanguageModelInput, Union[Dict, BaseModel]]:
+        """Model wrapper that returns outputs formatted to match the given schema.
+
+        Args:
+            schema:
+                The output schema. Can be passed in as:
+
+                - a Pydantic class,
+                - a JSON schema
+                - a TypedDict class
+                - an OpenAI function/tool schema.
+
+                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.
+
+            method: The method for steering model generation, one of:
+
+                - "function_calling":
+                    Uses Ollama's tool-calling API
+                - "json_schema":
+                    Uses Ollama's structured output API: https://ollama.com/blog/structured-outputs
+                - "json_mode":
+                    Specifies ``format="json"``. Note that if using JSON mode then you
+                    must include instructions for formatting the output into the
+                    desired schema into the model call.
+
+            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
+                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".
+
+            kwargs: Additional keyword args aren't supported.
+
+        Returns:
+            A Runnable that takes same inputs as a :class:`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:
+
+            - "raw": BaseMessage
+            - "parsed": None if there was a parsing error, otherwise the type depends on the ``schema`` as described above.
+            - "parsing_error": Optional[BaseException]
+
+        .. versionchanged:: 0.2.2
+
+            Added support for structured output API via ``format`` parameter.
+
+        .. dropdown:: Example: schema=Pydantic class, method="function_calling", include_raw=False
+
+            .. code-block:: python
+
+                from typing import Optional
+
+                from langchain_ollama import ChatOllama
+                from pydantic import BaseModel, Field
+
+
+                class AnswerWithJustification(BaseModel):
+                    '''An answer to the user question along with justification for the answer.'''
+
+                    answer: str
+                    justification: Optional[str] = Field(
+                        default=..., description="A justification for the answer."
+                    )
+
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(
+                    AnswerWithJustification
+                )
+
+                structured_llm.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.'
+                # )
+
+        .. dropdown:: Example: schema=Pydantic class, method="function_calling", include_raw=True
+
+            .. code-block:: python
+
+                from langchain_ollama import ChatOllama
+                from pydantic import BaseModel
+
+
+                class AnswerWithJustification(BaseModel):
+                    '''An answer to the user question along with justification for the answer.'''
+
+                    answer: str
+                    justification: str
+
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(
+                    AnswerWithJustification, include_raw=True
+                )
+
+                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
+                # }
+
+        .. dropdown:: Example: schema=Pydantic class, method="json_schema", include_raw=False
+
+            .. code-block:: python
+
+                from typing import Optional
+
+                from langchain_ollama import ChatOllama
+                from pydantic import BaseModel, Field
+
+
+                class AnswerWithJustification(BaseModel):
+                    '''An answer to the user question along with justification for the answer.'''
+
+                    answer: str
+                    justification: Optional[str] = Field(
+                        default=..., description="A justification for the answer."
+                    )
+
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(
+                    AnswerWithJustification, method="json_schema"
+                )
+
+                structured_llm.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.'
+                # )
+
+        .. dropdown:: Example: schema=TypedDict class, method="function_calling", include_raw=False
+
+            .. code-block:: python
+
+                # IMPORTANT: If you are using Python <=3.8, you need to import Annotated
+                # from typing_extensions, not from typing.
+                from typing_extensions import Annotated, TypedDict
+
+                from langchain_ollama import ChatOllama
+
+
+                class AnswerWithJustification(TypedDict):
+                    '''An answer to the user question along with justification for the answer.'''
+
+                    answer: str
+                    justification: Annotated[
+                        Optional[str], None, "A justification for the answer."
+                    ]
+
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(AnswerWithJustification)
+
+                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.'
+                # }
+
+        .. dropdown:: Example: schema=OpenAI function schema, method="function_calling", include_raw=False
+
+            .. code-block:: python
+
+                from langchain_ollama import ChatOllama
+
+                oai_schema = {
+                    'name': 'AnswerWithJustification',
+                    'description': 'An answer to the user question along with justification for the answer.',
+                    'parameters': {
+                        'type': 'object',
+                        'properties': {
+                            'answer': {'type': 'string'},
+                            'justification': {'description': 'A justification for the answer.', 'type': 'string'}
+                        },
+                       'required': ['answer']
+                   }
+               }
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(oai_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.'
+                # }
+
+        .. dropdown:: Example: schema=Pydantic class, method="json_mode", include_raw=True
+
+            .. code-block::
+
+                from langchain_ollama import ChatOllama
+                from pydantic import BaseModel
+
+                class AnswerWithJustification(BaseModel):
+                    answer: str
+                    justification: str
+
+                llm = ChatOllama(model="llama3.1", temperature=0)
+                structured_llm = llm.with_structured_output(
+                    AnswerWithJustification,
+                    method="json_mode",
+                    include_raw=True
+                )
+
+                structured_llm.invoke(
+                    "Answer the following question. "
+                    "Make sure to return a JSON blob with keys 'answer' and 'justification'.\\n\\n"
+                    "What's heavier a pound of bricks or a pound of feathers?"
+                )
+                # -> {
+                #     'raw': AIMessage(content='{\\n    "answer": "They are both the same weight.",\\n    "justification": "Both a pound of bricks and a pound of feathers weigh one pound. The difference lies in the volume and density of the materials, not the weight." \\n}'),
+                #     'parsed': AnswerWithJustification(answer='They are both the same weight.', justification='Both a pound of bricks and a pound of feathers weigh one pound. The difference lies in the volume and density of the materials, not the weight.'),
+                #     'parsing_error': None
+                # }
+        """  # noqa: E501, D301
+        if kwargs:
+            raise ValueError(f"Received unsupported arguments {kwargs}")
+        is_pydantic_schema = _is_pydantic_class(schema)
+        if method == "function_calling":
+            if schema is None:
+                raise ValueError(
+                    "schema must be specified when method is not 'json_mode'. "
+                    "Received None."
+                )
+            tool_name = convert_to_openai_tool(schema)["function"]["name"]
+            llm = self.bind_tools([schema], tool_choice=tool_name)
+            if is_pydantic_schema:
+                output_parser: Runnable = PydanticToolsParser(
+                    tools=[schema],  # type: ignore[list-item]
+                    first_tool_only=True,
+                )
+            else:
+                output_parser = JsonOutputKeyToolsParser(
+                    key_name=tool_name, first_tool_only=True
+                )
+        elif method == "json_mode":
+            llm = self.bind(format="json")
+            output_parser = (
+                PydanticOutputParser(pydantic_object=schema)  # type: ignore[arg-type]
+                if is_pydantic_schema
+                else JsonOutputParser()
+            )
+        elif method == "json_schema":
+            if schema is None:
+                raise ValueError(
+                    "schema must be specified when method is not 'json_mode'. "
+                    "Received None."
+                )
+            if is_pydantic_schema:
+                schema = cast(TypeBaseModel, schema)
+                llm = self.bind(format=schema.model_json_schema())
+                output_parser = PydanticOutputParser(pydantic_object=schema)
+            else:
+                if is_typeddict(schema):
+                    schema = cast(type, schema)
+                    response_format = convert_any_typed_dicts_to_pydantic(
+                        schema, visited={}
+                    ).schema()  # type: ignore[attr-defined]
+                    if "required" not in response_format:
+                        response_format["required"] = list(
+                            response_format["properties"].keys()
+                        )
+                else:
+                    # is JSON schema
+                    response_format = schema
+                llm = self.bind(format=response_format)
+                output_parser = JsonOutputParser()
+        else:
+            raise ValueError(
+                f"Unrecognized method argument. Expected one of 'function_calling', "
+                f"'json_schema', or 'json_mode'. Received: '{method}'"
+            )
+
+        if include_raw:
+            parser_assign = RunnablePassthrough.assign(
+                parsed=itemgetter("raw") | output_parser, parsing_error=lambda _: None
+            )
+            parser_none = RunnablePassthrough.assign(parsed=lambda _: None)
+            parser_with_fallback = parser_assign.with_fallbacks(
+                [parser_none], exception_key="parsing_error"
+            )
+            return RunnableMap(raw=llm) | parser_with_fallback
+        else:
+            return llm | output_parser

langchain_ollama/embeddings.py

@@ -1,3 +1,5 @@
+"""Ollama embeddings models."""
+
 from typing import (
     List,
     Optional,
@@ -132,12 +134,12 @@ class OllamaEmbeddings(BaseModel, Embeddings):
     For a full list of the params, see [this link](https://pydoc.dev/httpx/latest/httpx.Client.html)
     """
 
-    _client: Client = PrivateAttr(default=None)
+    _client: Client = PrivateAttr(default=None)  # type: ignore
     """
     The client to use for making requests.
     """
 
-    _async_client: AsyncClient = PrivateAttr(default=None)
+    _async_client: AsyncClient = PrivateAttr(default=None)  # type: ignore
     """
     The async client to use for making requests.
     """

langchain_ollama/llms.py

@@ -116,23 +116,30 @@ class OllamaLLM(BaseLLM):
     For a full list of the params, see [this link](https://pydoc.dev/httpx/latest/httpx.Client.html)
     """
 
-    _client: Client = PrivateAttr(default=None)
+    _client: Client = PrivateAttr(default=None)  # type: ignore
     """
     The client to use for making requests.
     """
 
-    _async_client: AsyncClient = PrivateAttr(default=None)
+    _async_client: AsyncClient = PrivateAttr(default=None)  # type: ignore
     """
     The async client to use for making requests.
     """
 
-    @property
-    def _default_params(self) -> Dict[str, Any]:
-        """Get the default parameters for calling Ollama."""
-        return {
-            "model": self.model,
-            "format": self.format,
-            "options": {
+    def _generate_params(
+        self,
+        prompt: str,
+        stop: Optional[List[str]] = None,
+        **kwargs: Any,
+    ) -> Dict[str, Any]:
+        if self.stop is not None and stop is not None:
+            raise ValueError("`stop` found in both the input and default params.")
+        elif self.stop is not None:
+            stop = self.stop
+
+        options_dict = kwargs.pop(
+            "options",
+            {
                 "mirostat": self.mirostat,
                 "mirostat_eta": self.mirostat_eta,
                 "mirostat_tau": self.mirostat_tau,
@@ -143,14 +150,25 @@ class OllamaLLM(BaseLLM):
                 "repeat_last_n": self.repeat_last_n,
                 "repeat_penalty": self.repeat_penalty,
                 "temperature": self.temperature,
-                "stop": self.stop,
+                "stop": self.stop if stop is None else stop,
                 "tfs_z": self.tfs_z,
                 "top_k": self.top_k,
                 "top_p": self.top_p,
             },
-            "keep_alive": self.keep_alive,
+        )
+
+        params = {
+            "prompt": prompt,
+            "stream": kwargs.pop("stream", True),
+            "model": kwargs.pop("model", self.model),
+            "format": kwargs.pop("format", self.format),
+            "options": Options(**options_dict),
+            "keep_alive": kwargs.pop("keep_alive", self.keep_alive),
+            **kwargs,
         }
 
+        return params
+
     @property
     def _llm_type(self) -> str:
         """Return type of LLM."""
@@ -179,27 +197,10 @@ class OllamaLLM(BaseLLM):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> AsyncIterator[Union[Mapping[str, Any], str]]:
-        if self.stop is not None and stop is not None:
-            raise ValueError("`stop` found in both the input and default params.")
-        elif self.stop is not None:
-            stop = self.stop
-
-        params = self._default_params
-
-        for key in self._default_params:
-            if key in kwargs:
-                params[key] = kwargs[key]
-
-        params["options"]["stop"] = stop
         async for part in await self._async_client.generate(
-            model=params["model"],
-            prompt=prompt,
-            stream=True,
-            options=Options(**params["options"]),
-            keep_alive=params["keep_alive"],
-            format=params["format"],
+            **self._generate_params(prompt, stop=stop, **kwargs)
         ):  # type: ignore
-            yield part
+            yield part  # type: ignore
 
     def _create_generate_stream(
         self,
@@ -207,26 +208,9 @@ class OllamaLLM(BaseLLM):
         stop: Optional[List[str]] = None,
         **kwargs: Any,
     ) -> Iterator[Union[Mapping[str, Any], str]]:
-        if self.stop is not None and stop is not None:
-            raise ValueError("`stop` found in both the input and default params.")
-        elif self.stop is not None:
-            stop = self.stop
-
-        params = self._default_params
-
-        for key in self._default_params:
-            if key in kwargs:
-                params[key] = kwargs[key]
-
-        params["options"]["stop"] = stop
         yield from self._client.generate(
-            model=params["model"],
-            prompt=prompt,
-            stream=True,
-            options=Options(**params["options"]),
-            keep_alive=params["keep_alive"],
-            format=params["format"],
-        )
+            **self._generate_params(prompt, stop=stop, **kwargs)
+        )  # type: ignore
 
     async def _astream_with_aggregation(
         self,

{langchain_ollama-0.2.0.dev1.dist-info → langchain_ollama-0.2.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: langchain-ollama
-Version: 0.2.0.dev1
+Version: 0.2.2
 Summary: An integration package connecting Ollama and LangChain
 Home-page: https://github.com/langchain-ai/langchain
 License: MIT
@@ -11,8 +11,9 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
-Requires-Dist: langchain-core (>=0.3.0.dev4,<0.4.0)
-Requires-Dist: ollama (>=0.3.0,<1)
+Classifier: Programming Language :: Python :: 3.13
+Requires-Dist: langchain-core (>=0.3.27,<0.4.0)
+Requires-Dist: ollama (>=0.4.4,<1)
 Project-URL: Repository, https://github.com/langchain-ai/langchain
 Project-URL: Release Notes, https://github.com/langchain-ai/langchain/releases?q=tag%3A%22langchain-ollama%3D%3D0%22&expanded=true
 Project-URL: Source Code, https://github.com/langchain-ai/langchain/tree/master/libs/partners/ollama

langchain_ollama-0.2.2.dist-info/RECORD

@@ -0,0 +1,9 @@
+langchain_ollama/__init__.py,sha256=SxPRrWcPayJpbwhheTtlqCaPp9ffiAAgZMM5Wc1yYpM,634
+langchain_ollama/chat_models.py,sha256=_kEf5o3CGtP7JL0HKZ8sIH9xdzktnGbBVdXULR2_sdo,47323
+langchain_ollama/embeddings.py,sha256=svqdPF44qX5qbFpZmLiXrzTC-AldmMlZRS5wBfY-EmA,5056
+langchain_ollama/llms.py,sha256=ojnYU0efhN10xhUINu1dCR2Erw79J_mYS6_l45J7Vls,12778
+langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+langchain_ollama-0.2.2.dist-info/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
+langchain_ollama-0.2.2.dist-info/METADATA,sha256=8eb8LP6TyzNBiiO60iS5ZMLKK0XW6PQUdKKQPHwYD7M,1876
+langchain_ollama-0.2.2.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+langchain_ollama-0.2.2.dist-info/RECORD,,

{langchain_ollama-0.2.0.dev1.dist-info → langchain_ollama-0.2.2.dist-info}/WHEEL

@@ -1,4 +1,4 @@
 Wheel-Version: 1.0
-Generator: poetry-core 1.8.1
+Generator: poetry-core 1.9.1
 Root-Is-Purelib: true
 Tag: py3-none-any

langchain_ollama-0.2.0.dev1.dist-info/RECORD

@@ -1,9 +0,0 @@
-langchain_ollama/__init__.py,sha256=HhQZqbCjhrbr2dC_9Dkw12pg4HPjnDXUoInROMNJKqA,518
-langchain_ollama/chat_models.py,sha256=q_URs_NzgY87XZ0RBDu-TY_seTh2lKXbtCXB7xY_utE,30423
-langchain_ollama/embeddings.py,sha256=46gmGxzK5Cm0GYesTSSgWupJYmJ2ywN7FQUAl0fzpxE,4991
-langchain_ollama/llms.py,sha256=uwQfKwDHXhWWVSAFzHpuv8SirBwKp0H4irnA8lqU0M4,13259
-langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-langchain_ollama-0.2.0.dev1.dist-info/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
-langchain_ollama-0.2.0.dev1.dist-info/METADATA,sha256=qI7Sy504_I0CEJJNrXBZwHTz1b_f6QMKYdjEowTETh4,1834
-langchain_ollama-0.2.0.dev1.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
-langchain_ollama-0.2.0.dev1.dist-info/RECORD,,