langchain-google-genai 2.1.8__tar.gz → 2.1.10__tar.gz

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: langchain-google-genai
-Version: 2.1.8
+Version: 2.1.10
 Summary: An integration package connecting Google's genai package and LangChain
 Home-page: https://github.com/langchain-ai/langchain-google
 License: MIT
@@ -13,7 +13,7 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Requires-Dist: filetype (>=1.2.0,<2.0.0)
 Requires-Dist: google-ai-generativelanguage (>=0.6.18,<0.7.0)
-Requires-Dist: langchain-core (>=0.3.68,<0.4.0)
+Requires-Dist: langchain-core (>=0.3.75,<0.4.0)
 Requires-Dist: pydantic (>=2,<3)
 Project-URL: Repository, https://github.com/langchain-ai/langchain-google
 Project-URL: Source Code, https://github.com/langchain-ai/langchain-google/tree/main/libs/genai
@@ -29,16 +29,20 @@ This package enables seamless access to Google Gemini's chat, vision, embeddings
 
 ## Table of Contents
 
-- [Overview](#overview)
-- [Installation](#installation)
-- [Quickstart](#quickstart)
-- [Chat Models](#chat-models)
-  - [Multimodal Inputs](#multimodal-inputs)
-  - [Multimodal Outputs](#multimodal-outputs)
-  - [Multimodal Outputs in Chains](#multimodal-outputs-in-chains)
-  - [Thinking Support](#thinking-support)
-- [Embeddings](#embeddings)
-- [Semantic Retrieval (RAG)](#semantic-retrieval-rag)
+- [langchain-google-genai](#langchain-google-genai)
+  - [Table of Contents](#table-of-contents)
+  - [Overview](#overview)
+  - [Installation](#installation)
+  - [Quickstart](#quickstart)
+  - [Chat Models](#chat-models)
+    - [Multimodal Inputs](#multimodal-inputs)
+    - [Multimodal Outputs](#multimodal-outputs)
+    - [Audio Output](#audio-output)
+    - [Multimodal Outputs in Chains](#multimodal-outputs-in-chains)
+    - [Thinking Support](#thinking-support)
+  - [Embeddings](#embeddings)
+  - [Semantic Retrieval (RAG)](#semantic-retrieval-rag)
+  - [Resources](#resources)
 
 ---
 
@@ -109,9 +113,9 @@ print(response.content)
 
 ✅ `image_url` can be:
 
-* A public image URL
-* A Google Cloud Storage path (`gcs://...`)
-* A base64-encoded image (e.g., `data:image/png;base64,...`)
+- A public image URL
+- A Google Cloud Storage path (`gcs://...`)
+- A base64-encoded image (e.g., `data:image/png;base64,...`)
 
 ---
 
@@ -208,7 +212,7 @@ You can use Gemini embeddings in LangChain:
 ```python
 from langchain_google_genai import GoogleGenerativeAIEmbeddings
 
-embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+embeddings = GoogleGenerativeAIEmbeddings(model="models/gemini-embedding-001")
 vector = embeddings.embed_query("hello, world!")
 print(vector)
 ```
@@ -249,12 +253,9 @@ print("Answerable probability:", response.answerable_probability)
 
 ---
 
-
 ## Resources
 
-* [LangChain Documentation](https://docs.langchain.com/)
-* [Google Generative AI SDK](https://googleapis.github.io/python-genai/)
-* [Gemini Model Documentation](https://ai.google.dev/)
-
-
+- [LangChain Documentation](https://docs.langchain.com/)
+- [Google Generative AI SDK](https://googleapis.github.io/python-genai/)
+- [Gemini Model Documentation](https://ai.google.dev/)
 

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/README.md

@@ -8,16 +8,20 @@ This package enables seamless access to Google Gemini's chat, vision, embeddings
 
 ## Table of Contents
 
-- [Overview](#overview)
-- [Installation](#installation)
-- [Quickstart](#quickstart)
-- [Chat Models](#chat-models)
-  - [Multimodal Inputs](#multimodal-inputs)
-  - [Multimodal Outputs](#multimodal-outputs)
-  - [Multimodal Outputs in Chains](#multimodal-outputs-in-chains)
-  - [Thinking Support](#thinking-support)
-- [Embeddings](#embeddings)
-- [Semantic Retrieval (RAG)](#semantic-retrieval-rag)
+- [langchain-google-genai](#langchain-google-genai)
+  - [Table of Contents](#table-of-contents)
+  - [Overview](#overview)
+  - [Installation](#installation)
+  - [Quickstart](#quickstart)
+  - [Chat Models](#chat-models)
+    - [Multimodal Inputs](#multimodal-inputs)
+    - [Multimodal Outputs](#multimodal-outputs)
+    - [Audio Output](#audio-output)
+    - [Multimodal Outputs in Chains](#multimodal-outputs-in-chains)
+    - [Thinking Support](#thinking-support)
+  - [Embeddings](#embeddings)
+  - [Semantic Retrieval (RAG)](#semantic-retrieval-rag)
+  - [Resources](#resources)
 
 ---
 
@@ -88,9 +92,9 @@ print(response.content)
 
 ✅ `image_url` can be:
 
-* A public image URL
-* A Google Cloud Storage path (`gcs://...`)
-* A base64-encoded image (e.g., `data:image/png;base64,...`)
+- A public image URL
+- A Google Cloud Storage path (`gcs://...`)
+- A base64-encoded image (e.g., `data:image/png;base64,...`)
 
 ---
 
@@ -187,7 +191,7 @@ You can use Gemini embeddings in LangChain:
 ```python
 from langchain_google_genai import GoogleGenerativeAIEmbeddings
 
-embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+embeddings = GoogleGenerativeAIEmbeddings(model="models/gemini-embedding-001")
 vector = embeddings.embed_query("hello, world!")
 print(vector)
 ```
@@ -228,11 +232,8 @@ print("Answerable probability:", response.answerable_probability)
 
 ---
 
-
 ## Resources
 
-* [LangChain Documentation](https://docs.langchain.com/)
-* [Google Generative AI SDK](https://googleapis.github.io/python-genai/)
-* [Gemini Model Documentation](https://ai.google.dev/)
-
-
+- [LangChain Documentation](https://docs.langchain.com/)
+- [Google Generative AI SDK](https://googleapis.github.io/python-genai/)
+- [Gemini Model Documentation](https://ai.google.dev/)

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/__init__.py

@@ -30,7 +30,7 @@ After setting up your environment with the required API key, you can interact wi
 
     from langchain_google_genai import ChatGoogleGenerativeAI
 
-    llm = ChatGoogleGenerativeAI(model="gemini-pro")
+    llm = ChatGoogleGenerativeAI(model="gemini-2.5-pro")
     llm.invoke("Sing a ballad of LangChain.")
 
 **Using LLMs**
@@ -41,7 +41,7 @@ The package also supports generating text with Google's models.
 
     from langchain_google_genai import GoogleGenerativeAI
 
-    llm = GoogleGenerativeAI(model="gemini-pro")
+    llm = GoogleGenerativeAI(model="gemini-2.5-pro")
     llm.invoke("Once upon a time, a library called LangChain")
 
 **Embedding Generation**
@@ -52,7 +52,7 @@ The package also supports creating embeddings with Google's models, useful for t
 
     from langchain_google_genai import GoogleGenerativeAIEmbeddings
 
-    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+    embeddings = GoogleGenerativeAIEmbeddings(model="models/gemini-embedding-001")
     embeddings.embed_query("hello, world!")
 
 """  # noqa: E501

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/_common.py

@@ -24,8 +24,8 @@ class _BaseGoogleGenerativeAI(BaseModel):
     model: str = Field(
         ...,
         description="""The name of the model to use.
-Supported examples:
-    - gemini-pro
+Examples:
+    - gemini-2.5-pro
     - models/text-bison-001""",
     )
     """Model name to use."""

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/_function_utils.py

@@ -322,7 +322,7 @@ def _get_properties_from_schema_any(schema: Any) -> Dict[str, Any]:
 
 
 def _get_properties_from_schema(schema: Dict) -> Dict[str, Any]:
-    properties = {}
+    properties: Dict[str, Dict[str, Union[str, int, Dict, List]]] = {}
     for k, v in schema.items():
         if not isinstance(k, str):
             logger.warning(f"Key '{k}' is not supported in schema, type={type(k)}")
@@ -331,7 +331,14 @@ def _get_properties_from_schema(schema: Dict) -> Dict[str, Any]:
             logger.warning(f"Value '{v}' is not supported in schema, ignoring v={v}")
             continue
         properties_item: Dict[str, Union[str, int, Dict, List]] = {}
-        if v.get("type") or v.get("anyOf") or v.get("type_"):
+        if v.get("anyOf") and all(
+            anyOf_type.get("type") != "null" for anyOf_type in v.get("anyOf", [])
+        ):
+            properties_item["anyOf"] = [
+                _format_json_schema_to_gapic(anyOf_type)
+                for anyOf_type in v.get("anyOf", [])
+            ]
+        elif v.get("type") or v.get("anyOf") or v.get("type_"):
             item_type_ = _get_type_from_schema(v)
             properties_item["type_"] = item_type_
             if _is_nullable_schema(v):

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/chat_models.py

@@ -6,6 +6,7 @@ import io
 import json
 import logging
 import mimetypes
+import time
 import uuid
 import warnings
 import wave
@@ -153,7 +154,12 @@ class ChatGoogleGenerativeAIError(GoogleGenerativeAIError):
     """
 
 
-def _create_retry_decorator() -> Callable[[Any], Any]:
+def _create_retry_decorator(
+    max_retries: int = 6,
+    wait_exponential_multiplier: float = 2.0,
+    wait_exponential_min: float = 1.0,
+    wait_exponential_max: float = 60.0,
+) -> Callable[[Any], Any]:
     """
     Creates and returns a preconfigured tenacity retry decorator.
 
@@ -165,15 +171,14 @@ def _create_retry_decorator() -> Callable[[Any], Any]:
         Callable[[Any], Any]: A retry decorator configured for handling specific
         Google API exceptions.
     """
-    multiplier = 2
-    min_seconds = 1
-    max_seconds = 60
-    max_retries = 2
-
     return retry(
         reraise=True,
         stop=stop_after_attempt(max_retries),
-        wait=wait_exponential(multiplier=multiplier, min=min_seconds, max=max_seconds),
+        wait=wait_exponential(
+            multiplier=wait_exponential_multiplier,
+            min=wait_exponential_min,
+            max=wait_exponential_max,
+        ),
         retry=(
             retry_if_exception_type(google.api_core.exceptions.ResourceExhausted)
             | retry_if_exception_type(google.api_core.exceptions.ServiceUnavailable)
@@ -198,13 +203,17 @@ def _chat_with_retry(generation_method: Callable, **kwargs: Any) -> Any:
     Returns:
         Any: The result from the chat generation method.
     """
-    retry_decorator = _create_retry_decorator()
+    retry_decorator = _create_retry_decorator(
+        max_retries=kwargs.get("max_retries", 6),
+        wait_exponential_multiplier=kwargs.get("wait_exponential_multiplier", 2.0),
+        wait_exponential_min=kwargs.get("wait_exponential_min", 1.0),
+        wait_exponential_max=kwargs.get("wait_exponential_max", 60.0),
+    )
 
     @retry_decorator
     def _chat_with_retry(**kwargs: Any) -> Any:
         try:
             return generation_method(**kwargs)
-        # Do not retry for these errors.
         except google.api_core.exceptions.FailedPrecondition as exc:
             if "location is not supported" in exc.message:
                 error_msg = (
@@ -218,6 +227,13 @@ def _chat_with_retry(generation_method: Callable, **kwargs: Any) -> Any:
             raise ChatGoogleGenerativeAIError(
                 f"Invalid argument provided to Gemini: {e}"
             ) from e
+        except google.api_core.exceptions.ResourceExhausted as e:
+            # Handle quota-exceeded error with recommended retry delay
+            if hasattr(e, "retry_after") and e.retry_after < kwargs.get(
+                "wait_exponential_max", 60.0
+            ):
+                time.sleep(e.retry_after)
+            raise e
         except Exception as e:
             raise e
 
@@ -682,7 +698,9 @@ def _parse_response_candidate(
                     )
     if content is None:
         content = ""
-    if any(isinstance(item, dict) and "executable_code" in item for item in content):
+    if isinstance(content, list) and any(
+        isinstance(item, dict) and "executable_code" in item for item in content
+    ):
         warnings.warn(
             """
         ⚠️ Warning: Output may vary each run.  
@@ -830,7 +848,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
 
             from langchain_google_genai import ChatGoogleGenerativeAI
 
-            llm = ChatGoogleGenerativeAI(model="gemini-2.0-flash-001")
+            llm = ChatGoogleGenerativeAI(model="gemini-2.5-flash")
             llm.invoke("Write me a ballad about LangChain")
 
     Invoke:
@@ -1040,7 +1058,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
         .. code-block:: python
 
             from google.ai.generativelanguage_v1beta.types import Tool as GenAITool
-            llm = ChatGoogleGenerativeAI(model="gemini-2.0-flash-exp")
+            llm = ChatGoogleGenerativeAI(model="gemini-2.5-flash")
             resp = llm.invoke(
                 "When is the next total solar eclipse in US?",
                 tools=[GenAITool(google_search={})],
@@ -1098,6 +1116,144 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
 
             'The weather in this image appears to be sunny and pleasant. The sky is a bright blue with scattered white clouds, suggesting fair weather. The lush green grass and trees indicate a warm and possibly slightly breezy day. There are no signs of rain or storms.'
 
+    PDF input:
+        .. code-block:: python
+
+            import base64
+            from langchain_core.messages import HumanMessage
+
+            pdf_bytes = open("/path/to/your/test.pdf", 'rb').read()
+            pdf_base64 = base64.b64encode(pdf_bytes).decode('utf-8')
+
+            message = HumanMessage(
+                content=[
+                    {"type": "text", "text": "describe the document in a sentence"},
+                    {
+                        "type": "file",
+                        "source_type": "base64",
+                        "mime_type":"application/pdf",
+                        "data": pdf_base64
+                    }
+                ]
+            )
+            ai_msg = llm.invoke([message])
+            ai_msg.content
+
+        .. code-block:: python
+
+            'This research paper describes a system developed for SemEval-2025 Task 9, which aims to automate the detection of food hazards from recall reports, addressing the class imbalance problem by leveraging LLM-based data augmentation techniques and transformer-based models to improve performance.'
+
+    Video input:
+        .. code-block:: python
+
+            import base64
+            from langchain_core.messages import HumanMessage
+
+            video_bytes = open("/path/to/your/video.mp4", 'rb').read()
+            video_base64 = base64.b64encode(video_bytes).decode('utf-8')
+
+            message = HumanMessage(
+                content=[
+                    {"type": "text", "text": "describe what's in this video in a sentence"},
+                    {
+                        "type": "file",
+                        "source_type": "base64",
+                        "mime_type": "video/mp4",
+                        "data": video_base64
+                    }
+                ]
+            )
+            ai_msg = llm.invoke([message])
+            ai_msg.content
+
+        .. code-block:: python
+
+            'Tom and Jerry, along with a turkey, engage in a chaotic Thanksgiving-themed adventure involving a corn-on-the-cob chase, maze antics, and a disastrous attempt to prepare a turkey dinner.'
+
+        You can also pass YouTube URLs directly:
+
+        .. code-block:: python
+
+            from langchain_core.messages import HumanMessage
+
+            message = HumanMessage(
+                content=[
+                    {"type": "text", "text": "summarize the video in 3 sentences."},
+                    {
+                        "type": "media",
+                        "file_uri": "https://www.youtube.com/watch?v=9hE5-98ZeCg",
+                        "mime_type": "video/mp4",
+                    }
+                ]
+            )
+            ai_msg = llm.invoke([message])
+            ai_msg.content
+
+        .. code-block:: python
+
+            'The video is a demo of multimodal live streaming in Gemini 2.0. The narrator is sharing his screen in AI Studio and asks if the AI can see it. The AI then reads text that is highlighted on the screen, defines the word “multimodal,” and summarizes everything that was seen and heard.'
+
+    Audio input:
+        .. code-block:: python
+
+            import base64
+            from langchain_core.messages import HumanMessage
+
+            audio_bytes = open("/path/to/your/audio.mp3", 'rb').read()
+            audio_base64 = base64.b64encode(audio_bytes).decode('utf-8')
+
+            message = HumanMessage(
+                content=[
+                    {"type": "text", "text": "summarize this audio in a sentence"},
+                    {
+                        "type": "file",
+                        "source_type": "base64",
+                        "mime_type":"audio/mp3",
+                        "data": audio_base64
+                    }
+                ]
+            )
+            ai_msg = llm.invoke([message])
+            ai_msg.content
+
+        .. code-block:: python
+
+            "In this episode of the Made by Google podcast, Stephen Johnson and Simon Tokumine discuss NotebookLM, a tool designed to help users understand complex material in various modalities, with a focus on its unexpected uses, the development of audio overviews, and the implementation of new features like mind maps and source discovery."
+
+    File upload (URI-based):
+        You can also upload files to Google's servers and reference them by URI.
+        This works for PDFs, images, videos, and audio files.
+
+        .. code-block:: python
+
+            import time
+            from google import genai
+            from langchain_core.messages import HumanMessage
+
+            client = genai.Client()
+
+            myfile = client.files.upload(file="/path/to/your/sample.pdf")
+            while myfile.state.name == "PROCESSING":
+                time.sleep(2)
+                myfile = client.files.get(name=myfile.name)
+
+            message = HumanMessage(
+                content=[
+                    {"type": "text", "text": "What is in the document?"},
+                    {
+                        "type": "media",
+                        "file_uri": myfile.uri,
+                        "mime_type": "application/pdf",
+                    },
+                ]
+            )
+            ai_msg = llm.invoke([message])
+            ai_msg.content
+
+        .. code-block:: python
+
+            "This research paper assesses and mitigates multi-turn jailbreak vulnerabilities in large language models using the Crescendo attack study, evaluating attack success rates and mitigation strategies like prompt hardening and LLM-as-guardrail."
+
     Token usage:
         .. code-block:: python
 
@@ -1127,8 +1283,8 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
 
     client: Any = Field(default=None, exclude=True)  #: :meta private:
     async_client_running: Any = Field(default=None, exclude=True)  #: :meta private:
-    default_metadata: Sequence[Tuple[str, str]] = Field(
-        default_factory=list
+    default_metadata: Optional[Sequence[Tuple[str, str]]] = Field(
+        default=None, alias="default_metadata_input"
     )  #: :meta private:
 
     convert_system_message_to_human: bool = False
@@ -1545,7 +1701,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
             )
 
             if run_manager:
-                run_manager.on_llm_new_token(gen.text)
+                run_manager.on_llm_new_token(gen.text, chunk=gen)
             yield gen
 
     async def _astream(
@@ -1611,7 +1767,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
                 )
 
                 if run_manager:
-                    await run_manager.on_llm_new_token(gen.text)
+                    await run_manager.on_llm_new_token(gen.text, chunk=gen)
                 yield gen
 
     def _prepare_request(
@@ -1816,7 +1972,8 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
             tools: A list of tool definitions to bind to this chat model.
                 Can be a pydantic model, callable, or BaseTool. Pydantic
                 models, callables, and BaseTools will be automatically converted to
-                their schema dictionary representation.
+                their schema dictionary representation. Tools with Union types in
+                their arguments are now supported and converted to `anyOf` schemas.
             **kwargs: Any additional parameters to pass to the
                 :class:`~langchain.runnable.Runnable` constructor.
         """

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/embeddings.py

@@ -40,7 +40,7 @@ class GoogleGenerativeAIEmbeddings(BaseModel, Embeddings):
 
             from langchain_google_genai import GoogleGenerativeAIEmbeddings
 
-            embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+            embeddings = GoogleGenerativeAIEmbeddings(model="gemini-embedding-001")
             embeddings.embed_query("What's our Q1 revenue?")
     """
 
@@ -49,7 +49,7 @@ class GoogleGenerativeAIEmbeddings(BaseModel, Embeddings):
     model: str = Field(
         ...,
         description="The name of the embedding model to use. "
-        "Example: ``'models/embedding-001'``",
+        "Example: ``'models/gemini-embedding-001'``",
     )
     task_type: Optional[str] = Field(
         default=None,

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/langchain_google_genai/llms.py

@@ -29,7 +29,7 @@ class GoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseLLM):
         .. code-block:: python
 
             from langchain_google_genai import GoogleGenerativeAI
-            llm = GoogleGenerativeAI(model="gemini-pro")
+            llm = GoogleGenerativeAI(model="gemini-2.5-pro")
     """
 
     client: Any = None  #: :meta private:

{langchain_google_genai-2.1.8 → langchain_google_genai-2.1.10}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "langchain-google-genai"
-version = "2.1.8"
+version = "2.1.10"
 description = "An integration package connecting Google's genai package and LangChain"
 authors = []
 readme = "README.md"
@@ -12,7 +12,7 @@ license = "MIT"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"
-langchain-core = "^0.3.68"
+langchain-core = "^0.3.75"
 google-ai-generativelanguage = "^0.6.18"
 pydantic = ">=2,<3"
 filetype = "^1.2.0"
@@ -23,7 +23,7 @@ optional = true
 [tool.poetry.group.test.dependencies]
 pytest = "^7.3.0"
 freezegun = "^1.2.2"
-pytest-mock = "^3.10.0"
+pytest-mock = "^3.14.1"
 syrupy = "^4.0.2"
 pytest-watcher = "^0.3.4"
 pytest-asyncio = "^0.21.1"
@@ -53,11 +53,11 @@ pytest = "^7.3.0"
 optional = true
 
 [tool.poetry.group.lint.dependencies]
-ruff = "^0.1.5"
+ruff = "^0.12.10"
 
 
 [tool.poetry.group.typing.dependencies]
-mypy = "^1.10"
+mypy = "^1.17.1"
 types-requests = "^2.31.0"
 types-google-cloud-ndb = "^2.2.0.1"
 types-protobuf = "^4.24.0.20240302"