langchain-ollama 0.1.1__py3-none-any.whl → 0.1.3__py3-none-any.whl

langchain_ollama/chat_models.py

@@ -35,6 +35,7 @@ from langchain_core.messages import (
 from langchain_core.messages.ai import UsageMetadata
 from langchain_core.messages.tool import tool_call
 from langchain_core.outputs import ChatGeneration, ChatGenerationChunk, ChatResult
+from langchain_core.pydantic_v1 import Field, root_validator
 from langchain_core.runnables import Runnable
 from langchain_core.tools import BaseTool
 from langchain_core.utils.function_calling import convert_to_openai_tool
@@ -322,6 +323,21 @@ class ChatOllama(BaseChatModel):
     base_url: Optional[str] = None
     """Base url the model is hosted under."""
 
+    client_kwargs: Optional[dict] = {}
+    """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 = Field(default=None)
+    """
+    The client to use for making requests.
+    """
+
+    _async_client: AsyncClient = Field(default=None)
+    """
+    The async client to use for making requests.
+    """
+
     @property
     def _default_params(self) -> Dict[str, Any]:
         """Get the default parameters for calling Ollama."""
@@ -348,6 +364,15 @@ class ChatOllama(BaseChatModel):
             "keep_alive": self.keep_alive,
         }
 
+    @root_validator(pre=False, skip_on_failure=True)
+    def _set_clients(cls, values: dict) -> dict:
+        """Set clients to use for ollama."""
+        values["_client"] = Client(host=values["base_url"], **values["client_kwargs"])
+        values["_async_client"] = AsyncClient(
+            host=values["base_url"], **values["client_kwargs"]
+        )
+        return values
+
     def _convert_messages_to_ollama_messages(
         self, messages: List[BaseMessage]
     ) -> Sequence[Message]:
@@ -449,7 +474,7 @@ class ChatOllama(BaseChatModel):
 
         params["options"]["stop"] = stop
         if "tools" in kwargs:
-            yield await AsyncClient(host=self.base_url).chat(
+            yield await self._async_client.chat(
                 model=params["model"],
                 messages=ollama_messages,
                 stream=False,
@@ -459,7 +484,7 @@ class ChatOllama(BaseChatModel):
                 tools=kwargs["tools"],
             )  # type:ignore
         else:
-            async for part in await AsyncClient(host=self.base_url).chat(
+            async for part in await self._async_client.chat(
                 model=params["model"],
                 messages=ollama_messages,
                 stream=True,
@@ -487,7 +512,7 @@ class ChatOllama(BaseChatModel):
 
         params["options"]["stop"] = stop
         if "tools" in kwargs:
-            yield Client(host=self.base_url).chat(
+            yield self._client.chat(
                 model=params["model"],
                 messages=ollama_messages,
                 stream=False,
@@ -497,7 +522,7 @@ class ChatOllama(BaseChatModel):
                 tools=kwargs["tools"],
             )
         else:
-            yield from Client(host=self.base_url).chat(
+            yield from self._client.chat(
                 model=params["model"],
                 messages=ollama_messages,
                 stream=True,

langchain_ollama/embeddings.py

@@ -1,34 +1,158 @@
-from typing import List
+from typing import (
+    List,
+    Optional,
+)
 
-import ollama
 from langchain_core.embeddings import Embeddings
-from langchain_core.pydantic_v1 import BaseModel, Extra
-from ollama import AsyncClient
+from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
+from ollama import AsyncClient, Client
 
 
 class OllamaEmbeddings(BaseModel, Embeddings):
-    """OllamaEmbeddings embedding model.
+    """Ollama embedding model integration.
 
-    Example:
+    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 .
+
+        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).
+
+        To fetch a model from the Ollama model library use ``ollama pull <name-of-model>``.
+
+        For example, to pull the llama3 model:
+
+        .. code-block:: bash
+
+            ollama pull llama3
+
+        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
+
+        You can specify the exact version of the model of interest
+        as such ``ollama pull vicuna:13b-v1.5-16k-q4_0``.
+
+        To view pulled models:
+
+        .. code-block:: bash
+
+            ollama list
+
+        To start serving:
+
+        .. code-block:: bash
+
+            ollama serve
+
+        View the Ollama documentation for more commands.
+
+        .. code-block:: bash
+
+            ollama help
+
+    Install the langchain-ollama integration package:
+        .. code-block:: bash
+
+            pip install -U langchain_ollama
+
+    Key init args — completion params:
+        model: str
+            Name of Ollama model to use.
+        base_url: Optional[str]
+            Base url the model is hosted under.
+
+    See full list of supported init args and their descriptions in the params section.
+
+    Instantiate:
         .. code-block:: python
 
             from langchain_ollama import OllamaEmbeddings
 
-            embedder = OllamaEmbeddings(model="llama3")
-            embedder.embed_query("what is the place that jonathan worked at?")
-    """
+            embed = OllamaEmbeddings(
+                model="llama3"
+            )
+
+    Embed single text:
+        .. code-block:: python
+
+            input_text = "The meaning of life is 42"
+            vector = embed.embed_query(input_text)
+            print(vector[:3])
+
+        .. code-block:: python
+
+            [-0.024603435769677162, -0.007543657906353474, 0.0039630369283258915]
+
+    Embed multiple texts:
+        .. code-block:: python
+
+             input_texts = ["Document 1...", "Document 2..."]
+            vectors = embed.embed_documents(input_texts)
+            print(len(vectors))
+            # The first 3 coordinates for the first vector
+            print(vectors[0][:3])
+
+        .. code-block:: python
+
+            2
+            [-0.024603435769677162, -0.007543657906353474, 0.0039630369283258915]
+
+    Async:
+        .. code-block:: python
+
+            vector = await embed.aembed_query(input_text)
+           print(vector[:3])
+
+            # multiple:
+            # await embed.aembed_documents(input_texts)
+
+        .. code-block:: python
+
+            [-0.009100092574954033, 0.005071679595857859, -0.0029193938244134188]
+    """  # noqa: E501
 
     model: str
     """Model name to use."""
 
+    base_url: Optional[str] = None
+    """Base url the model is hosted under."""
+
+    client_kwargs: Optional[dict] = {}
+    """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 = Field(default=None)
+    """
+    The client to use for making requests.
+    """
+
+    _async_client: AsyncClient = Field(default=None)
+    """
+    The async client to use for making requests.
+    """
+
     class Config:
         """Configuration for this pydantic object."""
 
-        extra = Extra.forbid
+        extra = "forbid"
+
+    @root_validator(pre=False, skip_on_failure=True)
+    def _set_clients(cls, values: dict) -> dict:
+        """Set clients to use for ollama."""
+        values["_client"] = Client(host=values["base_url"], **values["client_kwargs"])
+        values["_async_client"] = AsyncClient(
+            host=values["base_url"], **values["client_kwargs"]
+        )
+        return values
 
     def embed_documents(self, texts: List[str]) -> List[List[float]]:
         """Embed search docs."""
-        embedded_docs = ollama.embed(self.model, texts)["embeddings"]
+        embedded_docs = self._client.embed(self.model, texts)["embeddings"]
         return embedded_docs
 
     def embed_query(self, text: str) -> List[float]:
@@ -37,7 +161,9 @@ class OllamaEmbeddings(BaseModel, Embeddings):
 
     async def aembed_documents(self, texts: List[str]) -> List[List[float]]:
         """Embed search docs."""
-        embedded_docs = (await AsyncClient().embed(self.model, texts))["embeddings"]
+        embedded_docs = (await self._async_client.embed(self.model, texts))[
+            "embeddings"
+        ]
         return embedded_docs
 
     async def aembed_query(self, text: str) -> List[float]:

langchain_ollama/llms.py

@@ -12,14 +12,14 @@ from typing import (
     Union,
 )
 
-import ollama
 from langchain_core.callbacks import (
     AsyncCallbackManagerForLLMRun,
     CallbackManagerForLLMRun,
 )
-from langchain_core.language_models import BaseLLM
+from langchain_core.language_models import BaseLLM, LangSmithParams
 from langchain_core.outputs import GenerationChunk, LLMResult
-from ollama import AsyncClient, Options
+from langchain_core.pydantic_v1 import Field, root_validator
+from ollama import AsyncClient, Client, Options
 
 
 class OllamaLLM(BaseLLM):
@@ -107,6 +107,24 @@ class OllamaLLM(BaseLLM):
     keep_alive: Optional[Union[int, str]] = None
     """How long the model will stay loaded into memory."""
 
+    base_url: Optional[str] = None
+    """Base url the model is hosted under."""
+
+    client_kwargs: Optional[dict] = {}
+    """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 = Field(default=None)
+    """
+    The client to use for making requests.
+    """
+
+    _async_client: AsyncClient = Field(default=None)
+    """
+    The async client to use for making requests.
+    """
+
     @property
     def _default_params(self) -> Dict[str, Any]:
         """Get the default parameters for calling Ollama."""
@@ -137,6 +155,24 @@ class OllamaLLM(BaseLLM):
         """Return type of LLM."""
         return "ollama-llm"
 
+    def _get_ls_params(
+        self, stop: Optional[List[str]] = None, **kwargs: Any
+    ) -> LangSmithParams:
+        """Get standard params for tracing."""
+        params = super()._get_ls_params(stop=stop, **kwargs)
+        if max_tokens := kwargs.get("num_predict", self.num_predict):
+            params["ls_max_tokens"] = max_tokens
+        return params
+
+    @root_validator(pre=False, skip_on_failure=True)
+    def _set_clients(cls, values: dict) -> dict:
+        """Set clients to use for ollama."""
+        values["_client"] = Client(host=values["base_url"], **values["client_kwargs"])
+        values["_async_client"] = AsyncClient(
+            host=values["base_url"], **values["client_kwargs"]
+        )
+        return values
+
     async def _acreate_generate_stream(
         self,
         prompt: str,
@@ -155,7 +191,7 @@ class OllamaLLM(BaseLLM):
                 params[key] = kwargs[key]
 
         params["options"]["stop"] = stop
-        async for part in await AsyncClient().generate(
+        async for part in await self._async_client.generate(
             model=params["model"],
             prompt=prompt,
             stream=True,
@@ -183,7 +219,7 @@ class OllamaLLM(BaseLLM):
                 params[key] = kwargs[key]
 
         params["options"]["stop"] = stop
-        yield from ollama.generate(
+        yield from self._client.generate(
             model=params["model"],
             prompt=prompt,
             stream=True,

{langchain_ollama-0.1.1.dist-info → langchain_ollama-0.1.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: langchain-ollama
-Version: 0.1.1
+Version: 0.1.3
 Summary: An integration package connecting Ollama and LangChain
 Home-page: https://github.com/langchain-ai/langchain
 License: MIT
@@ -11,7 +11,7 @@ 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 (>=0.2.20,<0.3.0)
+Requires-Dist: langchain-core (>=0.2.36,<0.3.0)
 Requires-Dist: ollama (>=0.3.0,<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

langchain_ollama-0.1.3.dist-info/RECORD

@@ -0,0 +1,9 @@
+langchain_ollama/__init__.py,sha256=HhQZqbCjhrbr2dC_9Dkw12pg4HPjnDXUoInROMNJKqA,518
+langchain_ollama/chat_models.py,sha256=M2q6WK8tvFN9Q_CVH_cAYvNbhx7HNtN2PIw5255cpxs,30451
+langchain_ollama/embeddings.py,sha256=Sq30Q2lWJxc5BT1TLqDgrH7AW3tEezhGAJw_5dkr_8s,5005
+langchain_ollama/llms.py,sha256=FObzYNIQJoE5kQY_sh12CGj8AQXiMkfCLqrzsM8VNE0,13269
+langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+langchain_ollama-0.1.3.dist-info/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
+langchain_ollama-0.1.3.dist-info/METADATA,sha256=MOgY4zrbTiykPqxq660QciasjgMkwZ8yWptXMmHd-Xw,1827
+langchain_ollama-0.1.3.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
+langchain_ollama-0.1.3.dist-info/RECORD,,

langchain_ollama-0.1.1.dist-info/RECORD

@@ -1,9 +0,0 @@
-langchain_ollama/__init__.py,sha256=HhQZqbCjhrbr2dC_9Dkw12pg4HPjnDXUoInROMNJKqA,518
-langchain_ollama/chat_models.py,sha256=1lRveHDQxtYz3nNgXNaPQUAuo7WTHTf6jZarenXeCuM,29642
-langchain_ollama/embeddings.py,sha256=8_vd3sZaBCHMGZFTjUtDRUWrmbXNyBZSazrAGtBfFSU,1371
-langchain_ollama/llms.py,sha256=XT8daS4gRx8_w4a3s2pMGP-G7kBMHSYcQBUq4GfAou4,11940
-langchain_ollama/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-langchain_ollama-0.1.1.dist-info/LICENSE,sha256=2btS8uNUDWD_UNjw9ba6ZJt_00aUjEw9CGyK-xIHY8c,1072
-langchain_ollama-0.1.1.dist-info/METADATA,sha256=F41rq4WnaphtdkyWuZkKnxEp3gRT70n_OSptvwQtYd4,1827
-langchain_ollama-0.1.1.dist-info/WHEEL,sha256=FMvqSimYX_P7y0a7UY-_Mc83r5zkBZsCYPm7Lr0Bsq4,88
-langchain_ollama-0.1.1.dist-info/RECORD,,