core,groq,openai,mistralai,robocorp,fireworks,anthropic[patch]: Update BaseModel subclass and instance checks to handle both v1 and proper namespaces (#24417)

After this PR chat models will correctly handle pydantic 2 with
bind_tools and with_structured_output.


```python
import pydantic
print(pydantic.__version__)
```
2.8.2

```python
from langchain_openai import ChatOpenAI
from pydantic import BaseModel, Field

class Add(BaseModel):
    x: int
    y: int

model = ChatOpenAI().bind_tools([Add])
print(model.invoke('2 + 5').tool_calls)

model = ChatOpenAI().with_structured_output(Add)
print(type(model.invoke('2 + 5')))
```

```
[{'name': 'Add', 'args': {'x': 2, 'y': 5}, 'id': 'call_PNUFa4pdfNOYXxIMHc6ps2Do', 'type': 'tool_call'}]
<class '__main__.Add'>
```


```python
from langchain_openai import ChatOpenAI
from pydantic.v1 import BaseModel, Field

class Add(BaseModel):
    x: int
    y: int

model = ChatOpenAI().bind_tools([Add])
print(model.invoke('2 + 5').tool_calls)

model = ChatOpenAI().with_structured_output(Add)
print(type(model.invoke('2 + 5')))
```

```python
[{'name': 'Add', 'args': {'x': 2, 'y': 5}, 'id': 'call_hhiHYP441cp14TtrHKx3Upg0', 'type': 'tool_call'}]
<class '__main__.Add'>
```

Addresses issues: https://github.com/langchain-ai/langchain/issues/22782

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>

libs/core/langchain_core/language_models/chat_models.py

@@ -55,11 +55,16 @@ from langchain_core.outputs import (
     RunInfo,
 )
 from langchain_core.prompt_values import ChatPromptValue, PromptValue, StringPromptValue
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
+from langchain_core.pydantic_v1 import (
+    BaseModel,
+    Field,
+    root_validator,
+)
 from langchain_core.runnables import RunnableMap, RunnablePassthrough
 from langchain_core.runnables.config import ensure_config, run_in_executor
 from langchain_core.tracers._streaming import _StreamingCallbackHandler
 from langchain_core.utils.function_calling import convert_to_openai_tool
+from langchain_core.utils.pydantic import is_basemodel_subclass
 
 if TYPE_CHECKING:
     from langchain_core.output_parsers.base import OutputParserLike
@@ -1162,7 +1167,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 "with_structured_output is not implemented for this model."
             )
         llm = self.bind_tools([schema], tool_choice="any")
-        if isinstance(schema, type) and issubclass(schema, BaseModel):
+        if isinstance(schema, type) and is_basemodel_subclass(schema):
             output_parser: OutputParserLike = PydanticToolsParser(
                 tools=[schema], first_tool_only=True
             )

libs/core/langchain_core/runnables/base.py

@@ -82,6 +82,7 @@ from langchain_core.runnables.utils import (
 )
 from langchain_core.utils.aiter import aclosing, atee, py_anext
 from langchain_core.utils.iter import safetee
+from langchain_core.utils.pydantic import is_basemodel_subclass
 
 if TYPE_CHECKING:
     from langchain_core.callbacks.manager import (
@@ -300,7 +301,7 @@ class Runnable(Generic[Input, Output], ABC):
         """
         root_type = self.InputType
 
-        if inspect.isclass(root_type) and issubclass(root_type, BaseModel):
+        if inspect.isclass(root_type) and is_basemodel_subclass(root_type):
             return root_type
 
         return create_model(
@@ -332,7 +333,7 @@ class Runnable(Generic[Input, Output], ABC):
         """
         root_type = self.OutputType
 
-        if inspect.isclass(root_type) and issubclass(root_type, BaseModel):
+        if inspect.isclass(root_type) and is_basemodel_subclass(root_type):
             return root_type
 
         return create_model(

libs/core/langchain_core/runnables/graph.py

@@ -22,6 +22,7 @@ from typing import (
 from uuid import UUID, uuid4
 
 from langchain_core.pydantic_v1 import BaseModel
+from langchain_core.utils.pydantic import is_basemodel_subclass
 
 if TYPE_CHECKING:
     from langchain_core.runnables.base import Runnable as RunnableType
@@ -229,7 +230,7 @@ def node_data_json(
                 "name": node_data_str(node.id, node.data),
             },
         }
-    elif inspect.isclass(node.data) and issubclass(node.data, BaseModel):
+    elif inspect.isclass(node.data) and is_basemodel_subclass(node.data):
         json = (
             {
                 "type": "schema",

libs/core/langchain_core/utils/function_calling.py

@@ -28,6 +28,7 @@ from langchain_core.messages import (
 )
 from langchain_core.pydantic_v1 import BaseModel
 from langchain_core.utils.json_schema import dereference_refs
+from langchain_core.utils.pydantic import is_basemodel_subclass
 
 if TYPE_CHECKING:
     from langchain_core.tools import BaseTool
@@ -100,7 +101,11 @@ def convert_pydantic_to_openai_function(
     Returns:
         The function description.
     """
-    schema = dereference_refs(model.schema())
+    if hasattr(model, "model_json_schema"):
+        schema = model.model_json_schema()  # Pydantic 2
+    else:
+        schema = model.schema()  # Pydantic 1
+    schema = dereference_refs(schema)
     schema.pop("definitions", None)
     title = schema.pop("title", "")
     default_description = schema.pop("description", "")
@@ -272,7 +277,7 @@ def convert_to_openai_function(
             "description": function.pop("description"),
             "parameters": function,
         }
-    elif isinstance(function, type) and issubclass(function, BaseModel):
+    elif isinstance(function, type) and is_basemodel_subclass(function):
         return cast(Dict, convert_pydantic_to_openai_function(function))
     elif isinstance(function, BaseTool):
         return cast(Dict, format_tool_to_openai_function(function))

libs/core/tests/unit_tests/prompts/test_structured.py

@@ -8,12 +8,13 @@ from langchain_core.load.load import loads
 from langchain_core.prompts.structured import StructuredPrompt
 from langchain_core.pydantic_v1 import BaseModel
 from langchain_core.runnables.base import Runnable, RunnableLambda
+from langchain_core.utils.pydantic import is_basemodel_subclass
 
 
 def _fake_runnable(
     schema: Union[Dict, Type[BaseModel]], _: Any
 ) -> Union[BaseModel, Dict]:
-    if isclass(schema) and issubclass(schema, BaseModel):
+    if isclass(schema) and is_basemodel_subclass(schema):
         return schema(name="yo", value=42)
     else:
         params = cast(Dict, schema)["parameters"]