langchain-ollama 0.3.5__py3-none-any.whl → 0.3.7__py3-none-any.whl

langchain_ollama/__init__.py

@@ -1,6 +1,16 @@
 """This is the langchain_ollama package.
 
-It provides infrastructure for interacting with the Ollama service.
+Provides infrastructure for interacting with the `Ollama <https://ollama.com/>`__
+service.
+
+.. note::
+    **Newly added in 0.3.4:** ``validate_model_on_init`` param on all models.
+    This parameter allows you to validate the model exists in Ollama locally on
+    initialization. If set to ``True``, it will raise an error if the model does not
+    exist locally. This is useful for ensuring that the model is available before
+    attempting to use it, especially in environments where models may not be
+    pre-downloaded.
+
 """
 
 from importlib import metadata
@@ -10,6 +20,8 @@ from langchain_ollama.embeddings import OllamaEmbeddings
 from langchain_ollama.llms import OllamaLLM
 
 try:
+    if __package__ is None:
+        raise metadata.PackageNotFoundError
     __version__ = metadata.version(__package__)
 except metadata.PackageNotFoundError:
     # Case where package metadata is not available.

langchain_ollama/_utils.py

@@ -1,11 +1,11 @@
-"""Utility functions for validating Ollama models."""
+"""Utility function to validate Ollama models."""
 
 from httpx import ConnectError
 from ollama import Client, ResponseError
 
 
 def validate_model(client: Client, model_name: str) -> None:
-    """Validate that a model exists in the Ollama instance.
+    """Validate that a model exists in the local Ollama instance.
 
     Args:
         client: The Ollama client.
@@ -29,7 +29,10 @@ def validate_model(client: Client, model_name: str) -> None:
             )
             raise ValueError(msg)
     except ConnectError as e:
-        msg = "Failed to connect to Ollama. Please check that Ollama is downloaded, running and accessible. https://ollama.com/download"  # noqa: E501
+        msg = (
+            "Failed to connect to Ollama. Please check that Ollama is downloaded, "
+            "running and accessible. https://ollama.com/download"
+        )
         raise ValueError(msg) from e
     except ResponseError as e:
         msg = (

langchain_ollama/chat_models.py

@@ -2,7 +2,9 @@
 
 from __future__ import annotations
 
+import ast
 import json
+import logging
 from collections.abc import AsyncIterator, Iterator, Mapping, Sequence
 from operator import itemgetter
 from typing import (
@@ -57,6 +59,8 @@ from typing_extensions import Self, is_typeddict
 
 from ._utils import validate_model
 
+log = logging.getLogger(__name__)
+
 
 def _get_usage_metadata_from_generation_info(
     generation_info: Optional[Mapping[str, Any]],
@@ -77,33 +81,46 @@ def _get_usage_metadata_from_generation_info(
 
 def _parse_json_string(
     json_string: str,
+    *,
     raw_tool_call: dict[str, Any],
-    skip: bool,  # noqa: FBT001
+    skip: bool,
 ) -> Any:
     """Attempt to parse a JSON string for tool calling.
 
+    It first tries to use the standard ``json.loads``. If that fails, it falls
+    back to ``ast.literal_eval`` to safely parse Python literals, which is more
+    robust against models using single quotes or containing apostrophes.
+
     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.
+        skip: Whether to ignore parsing errors and return the value anyways.
 
     Returns:
-        The parsed JSON string.
+        The parsed JSON string or Python literal.
 
     Raises:
-        OutputParserException: If the JSON string wrong invalid and skip=False.
+        OutputParserException: If the string is 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 json.JSONDecodeError:
+        try:
+            # Use ast.literal_eval to safely parse Python-style dicts
+            # (e.g. with single quotes)
+            return ast.literal_eval(json_string)
+        except (SyntaxError, ValueError) as e:
+            # If both fail, and we're not skipping, raise an informative error.
+            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 or a Python literal. "
+                f"Received error: {e}"
+            )
+            raise OutputParserException(msg) from e
     except TypeError as e:
         if skip:
             return json_string
@@ -122,14 +139,20 @@ def _parse_arguments_from_tool_call(
 
     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
+    function_name = raw_tool_call["function"]["name"]
     arguments = raw_tool_call["function"]["arguments"]
     parsed_arguments: dict = {}
     if isinstance(arguments, dict):
         for key, value in arguments.items():
+            # Filter out metadata fields like 'functionName' that echo function name
+            if key == "functionName" and value == function_name:
+                continue
             if isinstance(value, str):
                 parsed_value = _parse_json_string(
                     value, skip=True, raw_tool_call=raw_tool_call
@@ -206,7 +229,7 @@ class ChatOllama(BaseChatModel):
 
         .. code-block:: bash
 
-            ollama pull mistral:v0.3
+            ollama pull gpt-oss:20b
             pip install -U langchain-ollama
 
     Key init args — completion params:
@@ -239,7 +262,8 @@ class ChatOllama(BaseChatModel):
             from langchain_ollama import ChatOllama
 
             llm = ChatOllama(
-                model = "llama3",
+                model = "gpt-oss:20b",
+                validate_model_on_init = True,
                 temperature = 0.8,
                 num_predict = 256,
                 # other params ...
@@ -261,10 +285,7 @@ class ChatOllama(BaseChatModel):
     Stream:
         .. code-block:: python
 
-            messages = [
-                ("human", "Return the words Hello World!"),
-            ]
-            for chunk in llm.stream(messages):
+            for chunk in llm.stream("Return the words Hello World!"):
                 print(chunk.text(), end="")
 
 
@@ -291,10 +312,7 @@ class ChatOllama(BaseChatModel):
     Async:
         .. code-block:: python
 
-            messages = [
-                ("human", "Hello how are you!"),
-            ]
-            await llm.ainvoke(messages)
+            await llm.ainvoke("Hello how are you!")
 
         .. code-block:: python
 
@@ -302,10 +320,7 @@ class ChatOllama(BaseChatModel):
 
         .. code-block:: python
 
-            messages = [
-                ("human", "Say hello world!"),
-            ]
-            async for chunk in llm.astream(messages):
+            async for chunk in llm.astream("Say hello world!"):
                 print(chunk.content)
 
         .. code-block:: python
@@ -333,10 +348,7 @@ class ChatOllama(BaseChatModel):
 
 
             json_llm = ChatOllama(format="json")
-            messages = [
-                ("human", "Return a query for the weather in a random location and time of day with two keys: location and time_of_day. Respond using JSON only."),
-            ]
-            llm.invoke(messages).content
+            llm.invoke("Return a query for the weather in a random location and time of day with two keys: location and time_of_day. Respond using JSON only.").content
 
         .. code-block:: python
 
@@ -383,17 +395,16 @@ class ChatOllama(BaseChatModel):
 
             llm = ChatOllama(
                 model = "deepseek-r1:8b",
+                validate_model_on_init = True,
                 reasoning= True,
             )
 
-            user_message = HumanMessage(content="how many r in the word strawberry?")
-            messages: List[Any] = [user_message]
-            llm.invoke(messages)
+            llm.invoke("how many r in the word strawberry?")
 
             # or, on an invocation basis:
 
-            llm.invoke(messages, reasoning=True)
-            # or llm.stream(messages, reasoning=True)
+            llm.invoke("how many r in the word strawberry?", reasoning=True)
+            # or llm.stream("how many r in the word strawberry?", reasoning=True)
 
             # If not provided, the invocation will default to the ChatOllama reasoning
             # param provided (None by default).
@@ -402,13 +413,12 @@ class ChatOllama(BaseChatModel):
 
             AIMessage(content='The word "strawberry" contains **three \'r\' letters**. Here\'s a breakdown for clarity:\n\n- The spelling of "strawberry" has two parts ... be 3.\n\nTo be thorough, let\'s confirm with an online source or common knowledge.\n\nI can recall that "strawberry" has: s-t-r-a-w-b-e-r-r-y — yes, three r\'s.\n\nPerhaps it\'s misspelled by some, but standard is correct.\n\nSo I think the response should be 3.\n'}, response_metadata={'model': 'deepseek-r1:8b', 'created_at': '2025-07-08T19:33:55.891269Z', 'done': True, 'done_reason': 'stop', 'total_duration': 98232561292, 'load_duration': 28036792, 'prompt_eval_count': 10, 'prompt_eval_duration': 40171834, 'eval_count': 3615, 'eval_duration': 98163832416, 'model_name': 'deepseek-r1:8b'}, id='run--18f8269f-6a35-4a7c-826d-b89d52c753b3-0', usage_metadata={'input_tokens': 10, 'output_tokens': 3615, 'total_tokens': 3625})
 
-
     """  # noqa: E501, pylint: disable=line-too-long
 
     model: str
     """Model name to use."""
 
-    reasoning: Optional[bool] = None
+    reasoning: Optional[Union[bool, str]] = None
     """Controls the reasoning/thinking mode for
     `supported models <https://ollama.com/search?c=thinking>`__.
 
@@ -421,33 +431,42 @@ class ChatOllama(BaseChatModel):
     - ``None`` (Default): The model will use its default reasoning behavior. Note
       however, if the model's default behavior *is* to perform reasoning, think tags
       ()``<think>`` and ``</think>``) will be present within the main response content
-      unless you set ``reasoning`` to ``True``."""
+      unless you set ``reasoning`` to ``True``.
+    - ``str``: e.g. ``'low'``, ``'medium'``, ``'high'``. Enables reasoning with a custom
+      intensity level. Currently, this is only supported ``gpt-oss``. See the
+      `Ollama docs <https://github.com/ollama/ollama-python/blob/da79e987f0ac0a4986bf396f043b36ef840370bc/ollama/_types.py#L210>`__
+      for more information.
+
+    """
 
     validate_model_on_init: bool = False
-    """Whether to validate the model exists in Ollama locally on initialization."""
+    """Whether to validate the model exists in Ollama locally on initialization.
+
+    .. versionadded:: 0.3.4
+    """
 
     mirostat: Optional[int] = None
     """Enable Mirostat sampling for controlling perplexity.
-    (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0)"""
+    (default: ``0``, ``0`` = disabled, ``1`` = Mirostat, ``2`` = Mirostat 2.0)"""
 
     mirostat_eta: Optional[float] = None
     """Influences how quickly the algorithm responds to feedback
     from the generated text. A lower learning rate will result in
     slower adjustments, while a higher learning rate will make
-    the algorithm more responsive. (Default: 0.1)"""
+    the algorithm more responsive. (Default: ``0.1``)"""
 
     mirostat_tau: Optional[float] = None
     """Controls the balance between coherence and diversity
     of the output. A lower value will result in more focused and
-    coherent text. (Default: 5.0)"""
+    coherent text. (Default: ``5.0``)"""
 
     num_ctx: Optional[int] = None
     """Sets the size of the context window used to generate the
-    next token. (Default: 2048)	"""
+    next token. (Default: ``2048``)	"""
 
     num_gpu: Optional[int] = None
-    """The number of GPUs to use. On macOS it defaults to 1 to
-    enable metal support, 0 to disable."""
+    """The number of GPUs to use. On macOS it defaults to ``1`` to
+    enable metal support, ``0`` to disable."""
 
     num_thread: Optional[int] = None
     """Sets the number of threads to use during computation.
@@ -457,20 +476,20 @@ class ChatOllama(BaseChatModel):
 
     num_predict: Optional[int] = None
     """Maximum number of tokens to predict when generating text.
-    (Default: 128, -1 = infinite generation, -2 = fill context)"""
+    (Default: ``128``, ``-1`` = infinite generation, ``-2`` = fill context)"""
 
     repeat_last_n: Optional[int] = None
     """Sets how far back for the model to look back to prevent
-    repetition. (Default: 64, 0 = disabled, -1 = num_ctx)"""
+    repetition. (Default: ``64``, ``0`` = disabled, ``-1`` = ``num_ctx``)"""
 
     repeat_penalty: Optional[float] = None
-    """Sets how strongly to penalize repetitions. A higher value (e.g., 1.5)
-    will penalize repetitions more strongly, while a lower value (e.g., 0.9)
-    will be more lenient. (Default: 1.1)"""
+    """Sets how strongly to penalize repetitions. A higher value (e.g., ``1.5``)
+    will penalize repetitions more strongly, while a lower value (e.g., ``0.9``)
+    will be more lenient. (Default: ``1.1``)"""
 
     temperature: Optional[float] = None
     """The temperature of the model. Increasing the temperature will
-    make the model answer more creatively. (Default: 0.8)"""
+    make the model answer more creatively. (Default: ``0.8``)"""
 
     seed: Optional[int] = None
     """Sets the random number seed to use for generation. Setting this
@@ -482,21 +501,21 @@ class ChatOllama(BaseChatModel):
 
     tfs_z: Optional[float] = None
     """Tail free sampling is used to reduce the impact of less probable
-    tokens from the output. A higher value (e.g., 2.0) will reduce the
-    impact more, while a value of 1.0 disables this setting. (default: 1)"""
+    tokens from the output. A higher value (e.g., ``2.0``) will reduce the
+    impact more, while a value of ``1.0`` disables this setting. (default: ``1``)"""
 
     top_k: Optional[int] = None
-    """Reduces the probability of generating nonsense. A higher value (e.g. 100)
-    will give more diverse answers, while a lower value (e.g. 10)
-    will be more conservative. (Default: 40)"""
+    """Reduces the probability of generating nonsense. A higher value (e.g. ``100``)
+    will give more diverse answers, while a lower value (e.g. ``10``)
+    will be more conservative. (Default: ``40``)"""
 
     top_p: Optional[float] = None
-    """Works together with top-k. A higher value (e.g., 0.95) will lead
-    to more diverse text, while a lower value (e.g., 0.5) will
-    generate more focused and conservative text. (Default: 0.9)"""
+    """Works together with top-k. A higher value (e.g., ``0.95``) will lead
+    to more diverse text, while a lower value (e.g., ``0.5``) will
+    generate more focused and conservative text. (Default: ``0.9``)"""
 
     format: Optional[Union[Literal["", "json"], JsonSchemaValue]] = None
-    """Specify the format of the output (options: "json", JSON schema)."""
+    """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."""
@@ -506,32 +525,35 @@ class ChatOllama(BaseChatModel):
 
     client_kwargs: Optional[dict] = {}
     """Additional kwargs to pass to the httpx clients.
+
     These arguments are passed to both synchronous and async clients.
-    Use sync_client_kwargs and async_client_kwargs to pass different arguments
+
+    Use ``sync_client_kwargs`` and ``async_client_kwargs`` to pass different arguments
     to synchronous and asynchronous clients.
+
     """
 
     async_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before
+    """Additional kwargs to merge with ``client_kwargs`` before
     passing to the httpx AsyncClient.
+
     `Full list of params. <https://www.python-httpx.org/api/#asyncclient>`__
+
     """
 
     sync_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before
+    """Additional kwargs to merge with ``client_kwargs`` before
     passing to the httpx Client.
+
     `Full list of params. <https://www.python-httpx.org/api/#client>`__
+
     """
 
     _client: Client = PrivateAttr()
-    """
-    The client to use for making requests.
-    """
+    """The client to use for making requests."""
 
     _async_client: AsyncClient = PrivateAttr()
-    """
-    The async client to use for making requests.
-    """
+    """The async client to use for making requests."""
 
     def _chat_params(
         self,
@@ -821,6 +843,28 @@ class ChatOllama(BaseChatModel):
         reasoning = kwargs.get("reasoning", self.reasoning)
         for stream_resp in self._create_chat_stream(messages, stop, **kwargs):
             if not isinstance(stream_resp, str):
+                content = (
+                    stream_resp["message"]["content"]
+                    if "message" in stream_resp and "content" in stream_resp["message"]
+                    else ""
+                )
+
+                # Warn and skip responses with done_reason: 'load' and empty content
+                # These indicate the model was loaded but no actual generation occurred
+                is_load_response_with_empty_content = (
+                    stream_resp.get("done") is True
+                    and stream_resp.get("done_reason") == "load"
+                    and not content.strip()
+                )
+
+                if is_load_response_with_empty_content:
+                    log.warning(
+                        "Ollama returned empty response with done_reason='load'."
+                        "This typically indicates the model was loaded but no content "
+                        "was generated. Skipping this response."
+                    )
+                    continue
+
                 if stream_resp.get("done") is True:
                     generation_info = dict(stream_resp)
                     if "model" in generation_info:
@@ -829,12 +873,6 @@ class ChatOllama(BaseChatModel):
                 else:
                     generation_info = None
 
-                content = (
-                    stream_resp["message"]["content"]
-                    if "message" in stream_resp and "content" in stream_resp["message"]
-                    else ""
-                )
-
                 additional_kwargs = {}
                 if (
                     reasoning
@@ -881,6 +919,28 @@ class ChatOllama(BaseChatModel):
         reasoning = kwargs.get("reasoning", self.reasoning)
         async for stream_resp in self._acreate_chat_stream(messages, stop, **kwargs):
             if not isinstance(stream_resp, str):
+                content = (
+                    stream_resp["message"]["content"]
+                    if "message" in stream_resp and "content" in stream_resp["message"]
+                    else ""
+                )
+
+                # Warn and skip responses with done_reason: 'load' and empty content
+                # These indicate the model was loaded but no actual generation occurred
+                is_load_response_with_empty_content = (
+                    stream_resp.get("done") is True
+                    and stream_resp.get("done_reason") == "load"
+                    and not content.strip()
+                )
+
+                if is_load_response_with_empty_content:
+                    log.warning(
+                        "Ollama returned empty response with done_reason='load'. "
+                        "This typically indicates the model was loaded but no content "
+                        "was generated. Skipping this response."
+                    )
+                    continue
+
                 if stream_resp.get("done") is True:
                     generation_info = dict(stream_resp)
                     if "model" in generation_info:
@@ -889,12 +949,6 @@ class ChatOllama(BaseChatModel):
                 else:
                     generation_info = None
 
-                content = (
-                    stream_resp["message"]["content"]
-                    if "message" in stream_resp and "content" in stream_resp["message"]
-                    else ""
-                )
-
                 additional_kwargs = {}
                 if (
                     reasoning
@@ -993,8 +1047,7 @@ class ChatOllama(BaseChatModel):
         """Model wrapper that returns outputs formatted to match the given schema.
 
         Args:
-            schema:
-                The output schema. Can be passed in as:
+            schema: The output schema. Can be passed in as:
 
                 - a Pydantic class,
                 - a JSON schema
@@ -1010,35 +1063,35 @@ class ChatOllama(BaseChatModel):
 
             method: The method for steering model generation, one of:
 
-                - "json_schema":
+                - ``'json_schema'``:
                     Uses Ollama's `structured output API <https://ollama.com/blog/structured-outputs>`__
-                - "function_calling":
+                - ``'function_calling'``:
                     Uses Ollama's tool-calling API
-                - "json_mode":
-                    Specifies ``format="json"``. Note that if using JSON mode then you
+                - ``'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
+                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".
+                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 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 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]
+            - ``'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
 
@@ -1046,7 +1099,7 @@ class ChatOllama(BaseChatModel):
 
         .. versionchanged:: 0.3.0
 
-            Updated default ``method`` to ``"json_schema"``.
+            Updated default ``method`` to ``'json_schema'``.
 
         .. dropdown:: Example: schema=Pydantic class, method="json_schema", include_raw=False
 
@@ -1081,7 +1134,7 @@ class ChatOllama(BaseChatModel):
                 #     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="json_schema", include_raw=True
+        .. dropdown:: Example: ``schema=Pydantic`` class, ``method='json_schema'``, ``include_raw=True``
 
             .. code-block:: python
 
@@ -1110,7 +1163,7 @@ class ChatOllama(BaseChatModel):
                 #     'parsing_error': None
                 # }
 
-        .. dropdown:: Example: schema=Pydantic class, method="function_calling", include_raw=False
+        .. dropdown:: Example: ``schema=Pydantic`` class, ``method='function_calling'``, ``include_raw=False``
 
             .. code-block:: python
 
@@ -1174,7 +1227,7 @@ class ChatOllama(BaseChatModel):
                 #     '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
+        .. dropdown:: Example: ``schema=OpenAI`` function schema, ``method='function_calling'``, ``include_raw=False``
 
             .. code-block:: python
 
@@ -1204,7 +1257,7 @@ class ChatOllama(BaseChatModel):
                 #     '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
+        .. dropdown:: Example: ``schema=Pydantic`` class, ``method='json_mode'``, ``include_raw=True``
 
             .. code-block::
 
@@ -1232,6 +1285,7 @@ class ChatOllama(BaseChatModel):
                 #     '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
         _ = kwargs.pop("strict", None)
         if kwargs:

langchain_ollama/embeddings.py

@@ -21,12 +21,12 @@ class OllamaEmbeddings(BaseModel, Embeddings):
     """Ollama embedding model integration.
 
     Set up a local Ollama instance:
-        Install the Ollama package and set up a local Ollama instance
-        using the instructions here: https://github.com/ollama/ollama .
+        `Install the Ollama package <https://github.com/ollama/ollama>`__ and set up a
+        local Ollama instance.
 
         You will need to choose a model to serve.
 
-        You can view a list of available models via the model library (https://ollama.com/library).
+        You can view a list of available models via `the model library <https://ollama.com/library>`__.
 
         To fetch a model from the Ollama model library use ``ollama pull <name-of-model>``.
 
@@ -39,8 +39,8 @@ class OllamaEmbeddings(BaseModel, Embeddings):
         This will download the default tagged version of the model.
         Typically, the default points to the latest, smallest sized-parameter model.
 
-        * On Mac, the models will be downloaded to ~/.ollama/models
-        * On Linux (or WSL), the models will be stored at /usr/share/ollama/.ollama/models
+        * On Mac, the models will be downloaded to ``~/.ollama/models``
+        * On Linux (or WSL), the models will be stored at ``/usr/share/ollama/.ollama/models``
 
         You can specify the exact version of the model of interest
         as such ``ollama pull vicuna:13b-v1.5-16k-q4_0``.
@@ -122,73 +122,80 @@ class OllamaEmbeddings(BaseModel, Embeddings):
         .. code-block:: python
 
             [-0.009100092574954033, 0.005071679595857859, -0.0029193938244134188]
+
     """  # noqa: E501
 
     model: str
     """Model name to use."""
 
     validate_model_on_init: bool = False
-    """Whether to validate the model exists in ollama locally on initialization."""
+    """Whether to validate the model exists in ollama locally on initialization.
+
+    .. versionadded:: 0.3.4
+
+    """
 
     base_url: Optional[str] = None
     """Base url the model is hosted under."""
 
     client_kwargs: Optional[dict] = {}
     """Additional kwargs to pass to the httpx clients.
+
     These arguments are passed to both synchronous and async clients.
-    Use sync_client_kwargs and async_client_kwargs to pass different arguments
+
+    Use ``sync_client_kwargs`` and ``async_client_kwargs`` to pass different arguments
     to synchronous and asynchronous clients.
+
     """
 
     async_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before passing to the httpx
+    """Additional kwargs to merge with ``client_kwargs`` before passing to the httpx
     AsyncClient.
 
     For a full list of the params, see the `HTTPX documentation <https://www.python-httpx.org/api/#asyncclient>`__.
+
     """
 
     sync_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before passing to the HTTPX Client.
+    """Additional kwargs to merge with ``client_kwargs`` before
+    passing to the HTTPX Client.
 
     For a full list of the params, see the `HTTPX documentation <https://www.python-httpx.org/api/#client>`__.
+
     """
 
     _client: Optional[Client] = PrivateAttr(default=None)
-    """
-    The client to use for making requests.
-    """
+    """The client to use for making requests."""
 
     _async_client: Optional[AsyncClient] = PrivateAttr(default=None)
-    """
-    The async client to use for making requests.
-    """
+    """The async client to use for making requests."""
 
     mirostat: Optional[int] = None
     """Enable Mirostat sampling for controlling perplexity.
-    (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0)"""
+    (default: ``0``, ``0`` = disabled, ``1`` = Mirostat, ``2`` = Mirostat 2.0)"""
 
     mirostat_eta: Optional[float] = None
     """Influences how quickly the algorithm responds to feedback
     from the generated text. A lower learning rate will result in
     slower adjustments, while a higher learning rate will make
-    the algorithm more responsive. (Default: 0.1)"""
+    the algorithm more responsive. (Default: ``0.1``)"""
 
     mirostat_tau: Optional[float] = None
     """Controls the balance between coherence and diversity
     of the output. A lower value will result in more focused and
-    coherent text. (Default: 5.0)"""
+    coherent text. (Default: ``5.0``)"""
 
     num_ctx: Optional[int] = None
     """Sets the size of the context window used to generate the
-    next token. (Default: 2048)	"""
+    next token. (Default: ``2048``)	"""
 
     num_gpu: Optional[int] = None
-    """The number of GPUs to use. On macOS it defaults to 1 to
-    enable metal support, 0 to disable."""
+    """The number of GPUs to use. On macOS it defaults to ``1`` to
+    enable metal support, ``0`` to disable."""
 
     keep_alive: Optional[int] = None
-    """controls how long the model will stay loaded into memory
-    following the request (default: 5m)
+    """Controls how long the model will stay loaded into memory
+    following the request (default: ``5m``)
     """
 
     num_thread: Optional[int] = None
@@ -199,34 +206,34 @@ class OllamaEmbeddings(BaseModel, Embeddings):
 
     repeat_last_n: Optional[int] = None
     """Sets how far back for the model to look back to prevent
-    repetition. (Default: 64, 0 = disabled, -1 = num_ctx)"""
+    repetition. (Default: ``64``, ``0`` = disabled, ``-1`` = ``num_ctx``)"""
 
     repeat_penalty: Optional[float] = None
-    """Sets how strongly to penalize repetitions. A higher value (e.g., 1.5)
-    will penalize repetitions more strongly, while a lower value (e.g., 0.9)
-    will be more lenient. (Default: 1.1)"""
+    """Sets how strongly to penalize repetitions. A higher value (e.g., ``1.5``)
+    will penalize repetitions more strongly, while a lower value (e.g., ``0.9``)
+    will be more lenient. (Default: ``1.1``)"""
 
     temperature: Optional[float] = None
     """The temperature of the model. Increasing the temperature will
-    make the model answer more creatively. (Default: 0.8)"""
+    make the model answer more creatively. (Default: ``0.8``)"""
 
     stop: Optional[list[str]] = None
     """Sets the stop tokens to use."""
 
     tfs_z: Optional[float] = None
     """Tail free sampling is used to reduce the impact of less probable
-    tokens from the output. A higher value (e.g., 2.0) will reduce the
-    impact more, while a value of 1.0 disables this setting. (default: 1)"""
+    tokens from the output. A higher value (e.g., ``2.0``) will reduce the
+    impact more, while a value of ``1.0`` disables this setting. (default: ``1``)"""
 
     top_k: Optional[int] = None
-    """Reduces the probability of generating nonsense. A higher value (e.g. 100)
-    will give more diverse answers, while a lower value (e.g. 10)
-    will be more conservative. (Default: 40)"""
+    """Reduces the probability of generating nonsense. A higher value (e.g. ``100``)
+    will give more diverse answers, while a lower value (e.g. ``10``)
+    will be more conservative. (Default: ``40``)"""
 
     top_p: Optional[float] = None
-    """Works together with top-k. A higher value (e.g., 0.95) will lead
-    to more diverse text, while a lower value (e.g., 0.5) will
-    generate more focused and conservative text. (Default: 0.9)"""
+    """Works together with top-k. A higher value (e.g., ``0.95``) will lead
+    to more diverse text, while a lower value (e.g., ``0.5``) will
+    generate more focused and conservative text. (Default: ``0.9``)"""
 
     model_config = ConfigDict(
         extra="forbid",
@@ -253,7 +260,7 @@ class OllamaEmbeddings(BaseModel, Embeddings):
 
     @model_validator(mode="after")
     def _set_clients(self) -> Self:
-        """Set clients to use for ollama."""
+        """Set clients to use for Ollama."""
         client_kwargs = self.client_kwargs or {}
 
         sync_client_kwargs = client_kwargs

langchain_ollama/llms.py

@@ -33,6 +33,7 @@ class OllamaLLM(BaseLLM):
 
             model = OllamaLLM(model="llama3")
             print(model.invoke("Come up with 10 names for a song about parrots"))
+
     """
 
     model: str
@@ -53,30 +54,33 @@ class OllamaLLM(BaseLLM):
       be present directly within the main response content."""
 
     validate_model_on_init: bool = False
-    """Whether to validate the model exists in ollama locally on initialization."""
+    """Whether to validate the model exists in ollama locally on initialization.
+
+    .. versionadded:: 0.3.4
+    """
 
     mirostat: Optional[int] = None
     """Enable Mirostat sampling for controlling perplexity.
-    (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0)"""
+    (default: ``0``, ``0`` = disabled, ``1`` = Mirostat, ``2`` = Mirostat 2.0)"""
 
     mirostat_eta: Optional[float] = None
     """Influences how quickly the algorithm responds to feedback
     from the generated text. A lower learning rate will result in
     slower adjustments, while a higher learning rate will make
-    the algorithm more responsive. (Default: 0.1)"""
+    the algorithm more responsive. (Default: ``0.1``)"""
 
     mirostat_tau: Optional[float] = None
     """Controls the balance between coherence and diversity
     of the output. A lower value will result in more focused and
-    coherent text. (Default: 5.0)"""
+    coherent text. (Default: ``5.0``)"""
 
     num_ctx: Optional[int] = None
     """Sets the size of the context window used to generate the
-    next token. (Default: 2048)"""
+    next token. (Default: ``2048``)"""
 
     num_gpu: Optional[int] = None
-    """The number of GPUs to use. On macOS it defaults to 1 to
-    enable metal support, 0 to disable."""
+    """The number of GPUs to use. On macOS it defaults to ``1`` to
+    enable metal support, ``0`` to disable."""
 
     num_thread: Optional[int] = None
     """Sets the number of threads to use during computation.
@@ -86,20 +90,20 @@ class OllamaLLM(BaseLLM):
 
     num_predict: Optional[int] = None
     """Maximum number of tokens to predict when generating text.
-    (Default: 128, -1 = infinite generation, -2 = fill context)"""
+    (Default: ``128``, ``-1`` = infinite generation, ``-2`` = fill context)"""
 
     repeat_last_n: Optional[int] = None
     """Sets how far back for the model to look back to prevent
-    repetition. (Default: 64, 0 = disabled, -1 = num_ctx)"""
+    repetition. (Default: ``64``, ``0`` = disabled, ``-1`` = ``num_ctx``)"""
 
     repeat_penalty: Optional[float] = None
-    """Sets how strongly to penalize repetitions. A higher value (e.g., 1.5)
-    will penalize repetitions more strongly, while a lower value (e.g., 0.9)
-    will be more lenient. (Default: 1.1)"""
+    """Sets how strongly to penalize repetitions. A higher value (e.g., ``1.5``)
+    will penalize repetitions more strongly, while a lower value (e.g., ``0.9``)
+    will be more lenient. (Default: ``1.1``)"""
 
     temperature: Optional[float] = None
     """The temperature of the model. Increasing the temperature will
-    make the model answer more creatively. (Default: 0.8)"""
+    make the model answer more creatively. (Default: ``0.8``)"""
 
     seed: Optional[int] = None
     """Sets the random number seed to use for generation. Setting this
@@ -111,21 +115,21 @@ class OllamaLLM(BaseLLM):
 
     tfs_z: Optional[float] = None
     """Tail free sampling is used to reduce the impact of less probable
-    tokens from the output. A higher value (e.g., 2.0) will reduce the
-    impact more, while a value of 1.0 disables this setting. (default: 1)"""
+    tokens from the output. A higher value (e.g., ``2.0``) will reduce the
+    impact more, while a value of 1.0 disables this setting. (default: ``1``)"""
 
     top_k: Optional[int] = None
-    """Reduces the probability of generating nonsense. A higher value (e.g. 100)
-    will give more diverse answers, while a lower value (e.g. 10)
-    will be more conservative. (Default: 40)"""
+    """Reduces the probability of generating nonsense. A higher value (e.g. ``100``)
+    will give more diverse answers, while a lower value (e.g. ``10``)
+    will be more conservative. (Default: ``40``)"""
 
     top_p: Optional[float] = None
-    """Works together with top-k. A higher value (e.g., 0.95) will lead
-    to more diverse text, while a lower value (e.g., 0.5) will
-    generate more focused and conservative text. (Default: 0.9)"""
+    """Works together with top-k. A higher value (e.g., ``0.95``) will lead
+    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)"""
+    """Specify the format of the output (options: ``'json'``)"""
 
     keep_alive: Optional[Union[int, str]] = None
     """How long the model will stay loaded into memory."""
@@ -135,33 +139,35 @@ class OllamaLLM(BaseLLM):
 
     client_kwargs: Optional[dict] = {}
     """Additional kwargs to pass to the httpx clients.
+
     These arguments are passed to both synchronous and async clients.
-    Use sync_client_kwargs and async_client_kwargs to pass different arguments
+
+    Use ``sync_client_kwargs`` and ``async_client_kwargs`` to pass different arguments
     to synchronous and asynchronous clients.
+
     """
 
     async_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before passing to the HTTPX
+    """Additional kwargs to merge with ``client_kwargs`` before passing to the HTTPX
     AsyncClient.
 
     For a full list of the params, see the `HTTPX documentation <https://www.python-httpx.org/api/#asyncclient>`__.
+
     """
 
     sync_client_kwargs: Optional[dict] = {}
-    """Additional kwargs to merge with client_kwargs before passing to the HTTPX Client.
+    """Additional kwargs to merge with ``client_kwargs`` before
+    passing to the HTTPX Client.
 
     For a full list of the params, see the `HTTPX documentation <https://www.python-httpx.org/api/#client>`__.
+
     """
 
     _client: Optional[Client] = PrivateAttr(default=None)
-    """
-    The client to use for making requests.
-    """
+    """The client to use for making requests."""
 
     _async_client: Optional[AsyncClient] = PrivateAttr(default=None)
-    """
-    The async client to use for making requests.
-    """
+    """The async client to use for making requests."""
 
     def _generate_params(
         self,

{langchain_ollama-0.3.5.dist-info → langchain_ollama-0.3.7.dist-info}/METADATA

@@ -1,14 +1,14 @@
 Metadata-Version: 2.1
 Name: langchain-ollama
-Version: 0.3.5
+Version: 0.3.7
 Summary: An integration package connecting Ollama and LangChain
 License: MIT
 Project-URL: Source Code, https://github.com/langchain-ai/langchain/tree/master/libs/partners/ollama
 Project-URL: Release Notes, https://github.com/langchain-ai/langchain/releases?q=tag%3A%22langchain-ollama%3D%3D0%22&expanded=true
 Project-URL: repository, https://github.com/langchain-ai/langchain
 Requires-Python: >=3.9
-Requires-Dist: ollama<1.0.0,>=0.5.1
-Requires-Dist: langchain-core<1.0.0,>=0.3.69
+Requires-Dist: ollama<1.0.0,>=0.5.3
+Requires-Dist: langchain-core<1.0.0,>=0.3.74
 Description-Content-Type: text/markdown
 
 # langchain-ollama

langchain_ollama-0.3.7.dist-info/RECORD

@@ -0,0 +1,11 @@
+langchain_ollama-0.3.7.dist-info/METADATA,sha256=JdFmlg4IUkM7Sy99773NdtIJKygkdJ6pYauIo29IWtQ,2072
+langchain_ollama-0.3.7.dist-info/WHEEL,sha256=9P2ygRxDrTJz3gsagc0Z96ukrxjr-LFBGOgv3AuKlCA,90
+langchain_ollama-0.3.7.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
+langchain_ollama-0.3.7.dist-info/licenses/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
+langchain_ollama/__init__.py,sha256=yXoECBZPpZGcg6V80qJCrgB02jW4KcjICZ5soLel6gw,1162
+langchain_ollama/_utils.py,sha256=amg8-DK0XaWe_aUO-ADFDUzb9EMJ-vMeyz2e1qrqAUo,1443
+langchain_ollama/chat_models.py,sha256=xV5R4hlOAQOsRwb4WdfR4UPveI3-QChBaa3nGa4Ny3I,58314
+langchain_ollama/embeddings.py,sha256=PxYDwvHbq-J6lTusSBIXtTeBnYSjG3R-a2wXFa52LXM,10455
+langchain_ollama/llms.py,sha256=65NARxA7xKYmwugvw4SxfpDPzHKiGGa6jsX7BgR64oM,16936
+langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+langchain_ollama-0.3.7.dist-info/RECORD,,

langchain_ollama-0.3.5.dist-info/RECORD

@@ -1,11 +0,0 @@
-langchain_ollama-0.3.5.dist-info/METADATA,sha256=15Fpg-jcUr1QRiPMxng3GPbvRHmAapo1s6qykTZvHk0,2072
-langchain_ollama-0.3.5.dist-info/WHEEL,sha256=9P2ygRxDrTJz3gsagc0Z96ukrxjr-LFBGOgv3AuKlCA,90
-langchain_ollama-0.3.5.dist-info/entry_points.txt,sha256=6OYgBcLyFCUgeqLgnvMyOJxPCWzgy7se4rLPKtNonMs,34
-langchain_ollama-0.3.5.dist-info/licenses/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
-langchain_ollama/__init__.py,sha256=TI1gI0Wpg7mRXehGpxrJG2flF_t4Ev-aIJlLKV-CgL0,633
-langchain_ollama/_utils.py,sha256=dmFO4tSvDTeMALc89QnTBLNWPMZL0eNAq1EDwuMjRA8,1416
-langchain_ollama/chat_models.py,sha256=olz3KJeLG1vk47Xl38nN9bP4bcol5cBQnPnu5MyP8k8,55539
-langchain_ollama/embeddings.py,sha256=walU1vZq_YamLLPDSJLbMtOu6jFbiNnhJ5ni2sybCRs,10318
-langchain_ollama/llms.py,sha256=PSJ-VQMocp1nm-pgtnKnozidt66RKJiEnhdzftoLNNc,16778
-langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-langchain_ollama-0.3.5.dist-info/RECORD,,