langchain-google-genai 2.1.0__tar.gz → 2.1.1__tar.gz

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: langchain-google-genai
-Version: 2.1.0
+Version: 2.1.1
 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 (>=0.6.16,<0.7.0)
-Requires-Dist: langchain-core (>=0.3.43,<0.4.0)
+Requires-Dist: langchain-core (>=0.3.47,<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
@@ -77,7 +77,44 @@ The value of `image_url` can be any of the following:
 - An accessible gcs file (e.g., "gcs://path/to/file.png")
 - A base64 encoded image (e.g., `data:image/png;base64,abcd124`)
 
+#### Multimodal outputs
 
+Gemini 2.0 Flash Experimental model supports text output with inline images
+
+```
+from langchain_google_genai import ChatGoogleGenerativeAI
+
+llm = ChatGoogleGenerativeAI(model="models/gemini-2.0-flash-exp-image-generation")
+# example
+response = llm.invoke(
+    "Generate an image of a cat and say meow",
+    generation_config=dict(response_modalities=["TEXT", "IMAGE"]),
+)
+
+# Base64 encoded binary data of the image
+image_base64 = response.content[0].get("image_url").get("url").split(",")[-1]
+meow_str = response.content[1]
+```
+
+#### Multimodal Outputs in Chains
+
+'''
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.prompts import ChatPromptTemplate
+
+from langchain_google_genai import ChatGoogleGenerativeAI, Modality
+
+llm = ChatGoogleGenerativeAI(
+    model="models/gemini-2.0-flash-exp-image-generation",
+    response_modalities=[Modality.TEXT, Modality.IMAGE],
+)
+
+prompt = ChatPromptTemplate(
+    [("human", "Generate an image of {animal} and tell me the sound of the animal")]
+)
+chain = {"animal": RunnablePassthrough()} | prompt | llm
+res = chain.invoke("cat")
+'''
 
 ## Embeddings
 

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/README.md

@@ -56,7 +56,44 @@ The value of `image_url` can be any of the following:
 - An accessible gcs file (e.g., "gcs://path/to/file.png")
 - A base64 encoded image (e.g., `data:image/png;base64,abcd124`)
 
+#### Multimodal outputs
 
+Gemini 2.0 Flash Experimental model supports text output with inline images
+
+```
+from langchain_google_genai import ChatGoogleGenerativeAI
+
+llm = ChatGoogleGenerativeAI(model="models/gemini-2.0-flash-exp-image-generation")
+# example
+response = llm.invoke(
+    "Generate an image of a cat and say meow",
+    generation_config=dict(response_modalities=["TEXT", "IMAGE"]),
+)
+
+# Base64 encoded binary data of the image
+image_base64 = response.content[0].get("image_url").get("url").split(",")[-1]
+meow_str = response.content[1]
+```
+
+#### Multimodal Outputs in Chains
+
+'''
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.prompts import ChatPromptTemplate
+
+from langchain_google_genai import ChatGoogleGenerativeAI, Modality
+
+llm = ChatGoogleGenerativeAI(
+    model="models/gemini-2.0-flash-exp-image-generation",
+    response_modalities=[Modality.TEXT, Modality.IMAGE],
+)
+
+prompt = ChatPromptTemplate(
+    [("human", "Generate an image of {animal} and tell me the sound of the animal")]
+)
+chain = {"animal": RunnablePassthrough()} | prompt | llm
+res = chain.invoke("cat")
+'''
 
 ## Embeddings
 

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/langchain_google_genai/__init__.py

@@ -55,7 +55,7 @@ embeddings.embed_query("hello, world!")
 ```
 """  # noqa: E501
 
-from langchain_google_genai._enums import HarmBlockThreshold, HarmCategory
+from langchain_google_genai._enums import HarmBlockThreshold, HarmCategory, Modality
 from langchain_google_genai.chat_models import ChatGoogleGenerativeAI
 from langchain_google_genai.embeddings import GoogleGenerativeAIEmbeddings
 from langchain_google_genai.genai_aqa import (
@@ -80,5 +80,6 @@ __all__ = [
     "GoogleVectorStore",
     "HarmBlockThreshold",
     "HarmCategory",
+    "Modality",
     "DoesNotExistsException",
 ]

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/langchain_google_genai/_common.py

@@ -1,11 +1,11 @@
 from importlib import metadata
-from typing import Any, Dict, Optional, Tuple, TypedDict
+from typing import Any, Dict, List, Optional, Tuple, TypedDict
 
 from google.api_core.gapic_v1.client_info import ClientInfo
 from langchain_core.utils import secret_from_env
 from pydantic import BaseModel, Field, SecretStr
 
-from langchain_google_genai._enums import HarmBlockThreshold, HarmCategory
+from langchain_google_genai._enums import HarmBlockThreshold, HarmCategory, Modality
 
 
 class GoogleGenerativeAIError(Exception):
@@ -72,6 +72,9 @@ Supported examples:
             "A key-value dictionary representing additional headers for the model call"
         ),
     )
+    response_modalities: Optional[List[Modality]] = Field(
+        default=None, description=("A list of modalities of the response")
+    )
 
     safety_settings: Optional[Dict[HarmCategory, HarmBlockThreshold]] = None
     """The default safety settings to use for all generations. 

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/langchain_google_genai/_enums.py

@@ -2,5 +2,6 @@ import google.ai.generativelanguage_v1beta as genai
 
 HarmBlockThreshold = genai.SafetySetting.HarmBlockThreshold
 HarmCategory = genai.HarmCategory
+Modality = genai.GenerationConfig.Modality
 
-__all__ = ["HarmBlockThreshold", "HarmCategory"]
+__all__ = ["HarmBlockThreshold", "HarmCategory", "Modality"]

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/langchain_google_genai/chat_models.py

@@ -106,7 +106,10 @@ from langchain_google_genai._function_utils import (
     is_basemodel_subclass_safe,
     tool_to_dict,
 )
-from langchain_google_genai._image_utils import ImageBytesLoader
+from langchain_google_genai._image_utils import (
+    ImageBytesLoader,
+    image_bytes_to_b64_string,
+)
 
 from . import _genai_extension as genaix
 
@@ -430,7 +433,7 @@ def _parse_chat_history(
 def _parse_response_candidate(
     response_candidate: Candidate, streaming: bool = False
 ) -> AIMessage:
-    content: Union[None, str, List[str]] = None
+    content: Union[None, str, List[Union[str, dict]]] = None
     additional_kwargs = {}
     tool_calls = []
     invalid_tool_calls = []
@@ -455,6 +458,26 @@ def _parse_response_candidate(
             elif text:
                 raise Exception("Unexpected content type")
 
+        if part.inline_data.mime_type.startswith("image/"):
+            image_format = part.inline_data.mime_type[6:]
+            message = {
+                "type": "image_url",
+                "image_url": {
+                    "url": image_bytes_to_b64_string(
+                        part.inline_data.data, image_format=image_format
+                    )
+                },
+            }
+
+            if not content:
+                content = [message]
+            elif isinstance(content, str) and message:
+                content = [content, message]
+            elif isinstance(content, list) and message:
+                content.append(message)
+            elif message:
+                raise Exception("Unexpected content type")
+
         if part.function_call:
             function_call = {"name": part.function_call.name}
             # dump to match other function calling llm for now
@@ -831,8 +854,8 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
     @model_validator(mode="after")
     def validate_environment(self) -> Self:
         """Validates params and passes them to google-generativeai package."""
-        if self.temperature is not None and not 0 <= self.temperature <= 1:
-            raise ValueError("temperature must be in the range [0.0, 1.0]")
+        if self.temperature is not None and not 0 <= self.temperature <= 2.0:
+            raise ValueError("temperature must be in the range [0.0, 2.0]")
 
         if self.top_p is not None and not 0 <= self.top_p <= 1:
             raise ValueError("top_p must be in the range [0.0, 1.0]")
@@ -895,6 +918,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
             "top_k": self.top_k,
             "n": self.n,
             "safety_settings": self.safety_settings,
+            "response_modalities": self.response_modalities,
         }
 
     def _get_ls_params(
@@ -928,6 +952,7 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
                 "max_output_tokens": self.max_output_tokens,
                 "top_k": self.top_k,
                 "top_p": self.top_p,
+                "response_modalities": self.response_modalities,
             }.items()
             if v is not None
         }
@@ -1256,6 +1281,8 @@ class ChatGoogleGenerativeAI(_BaseGoogleGenerativeAI, BaseChatModel):
         include_raw: bool = False,
         **kwargs: Any,
     ) -> Runnable[LanguageModelInput, Union[Dict, BaseModel]]:
+        _ = kwargs.pop("method", None)
+        _ = kwargs.pop("strict", None)
         if kwargs:
             raise ValueError(f"Received unsupported arguments {kwargs}")
         tool_name = _get_tool_name(schema)  # type: ignore[arg-type]

{langchain_google_genai-2.1.0 → langchain_google_genai-2.1.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "langchain-google-genai"
-version = "2.1.0"
+version = "2.1.1"
 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 = "^0.3.43"
+langchain-core = "^0.3.47"
 google-ai-generativelanguage = "^0.6.16"
 pydantic = ">=2,<3"
 filetype = "^1.2.0"
@@ -28,7 +28,7 @@ syrupy = "^4.0.2"
 pytest-watcher = "^0.3.4"
 pytest-asyncio = "^0.21.1"
 numpy = "^1.26.2"
-langchain-tests = "0.3.14"
+langchain-tests = "0.3.15"
 
 [tool.codespell]
 ignore-words-list = "rouge"