## Goal Solve the following problems with `langchain-openai`: - Structured output with `o1` [breaks out of the box](https://langchain.slack.com/archives/C050X0VTN56/p1735232400232099). - `with_structured_output` by default does not use OpenAI’s [structured output feature](https://platform.openai.com/docs/guides/structured-outputs). - We override API defaults for temperature and other parameters. ## Breaking changes: - Default method for structured output is changing to OpenAI’s dedicated [structured output feature](https://platform.openai.com/docs/guides/structured-outputs). For schemas specified via TypedDict or JSON schema, strict schema validation is disabled by default but can be enabled by specifying `strict=True`. - To recover previous default, pass `method="function_calling"` into `with_structured_output`. - Models that don’t support `method="json_schema"` (e.g., `gpt-4` and `gpt-3.5-turbo`, currently the default model for ChatOpenAI) will raise an error unless `method` is explicitly specified. - To recover previous default, pass `method="function_calling"` into `with_structured_output`. - Schemas specified via Pydantic `BaseModel` that have fields with non-null defaults or metadata (like min/max constraints) will raise an error. - To recover previous default, pass `method="function_calling"` into `with_structured_output`. - `strict` now defaults to False for `method="json_schema"` when schemas are specified via TypedDict or JSON schema. - To recover previous behavior, use `with_structured_output(schema, strict=True)` - Schemas specified via Pydantic V1 will raise a warning (and use `method="function_calling"`) unless `method` is explicitly specified. - To remove the warning, pass `method="function_calling"` into `with_structured_output`. - Streaming with default structured output method / Pydantic schema no longer generates intermediate streamed chunks. - To recover previous behavior, pass `method="function_calling"` into `with_structured_output`. - We no longer override default temperature (was 0.7 in LangChain, now will follow OpenAI, currently 1.0). - To recover previous behavior, initialize `ChatOpenAI` or `AzureChatOpenAI` with `temperature=0.7`. - Note: conceptually there is a difference between forcing a tool call and forcing a response format. Tool calls may have more concise arguments vs. generating content adhering to a schema. Prompts may need to be adjusted to recover desired behavior. --------- Co-authored-by: Jacob Lee <jacoblee93@gmail.com> Co-authored-by: Bagatur <baskaryan@gmail.com> |
||
---|---|---|
.. | ||
langchain_tests | ||
scripts | ||
tests | ||
Makefile | ||
poetry.lock | ||
pyproject.toml | ||
README.md |
langchain-tests
This is a testing library for LangChain integrations. It contains the base classes for a standard set of tests.
Installation
We encourage pinning your version to a specific version in order to avoid breaking your CI when we publish new tests. We recommend upgrading to the latest version periodically to make sure you have the latest tests.
Not pinning your version will ensure you always have the latest tests, but it may also break your CI if we introduce tests that your integration doesn't pass.
Pip:
```bash
pip install -U langchain-tests
```
Poetry:
```bash
poetry add langchain-tests
```
Usage
To add standard tests to an integration package's e.g. ChatModel, you need to create
- A unit test class that inherits from ChatModelUnitTests
- An integration test class that inherits from ChatModelIntegrationTests
tests/unit_tests/test_standard.py
:
"""Standard LangChain interface tests"""
from typing import Type
import pytest
from langchain_core.language_models import BaseChatModel
from langchain_tests.unit_tests import ChatModelUnitTests
from langchain_parrot_chain import ChatParrotChain
class TestParrotChainStandard(ChatModelUnitTests):
@pytest.fixture
def chat_model_class(self) -> Type[BaseChatModel]:
return ChatParrotChain
tests/integration_tests/test_standard.py
:
"""Standard LangChain interface tests"""
from typing import Type
import pytest
from langchain_core.language_models import BaseChatModel
from langchain_tests.integration_tests import ChatModelIntegrationTests
from langchain_parrot_chain import ChatParrotChain
class TestParrotChainStandard(ChatModelIntegrationTests):
@pytest.fixture
def chat_model_class(self) -> Type[BaseChatModel]:
return ChatParrotChain
Reference
The following fixtures are configurable in the test classes. Anything not marked as required is optional.
chat_model_class
(required): The class of the chat model to be testedchat_model_params
: The keyword arguments to pass to the chat model constructorchat_model_has_tool_calling
: Whether the chat model can call tools. By default, this is set tohasattr(chat_model_class, 'bind_tools)
chat_model_has_structured_output
: Whether the chat model can structured output. By default, this is set tohasattr(chat_model_class, 'with_structured_output')