ChatBedrockConverse: additional_model_request_fields nested keys incorrectly converted to snake_case during invoke()

[mshavliuk]

Package

langchain-aws

Checked other resources

• I added a descriptive title to this issue
• I searched the LangChain documentation with the integrated search
• I used the GitHub search to find a similar issue and didn't find it
• I am sure this is a bug and not a question or request for help

Example Code

import os

from langchain_aws import ChatBedrockConverse

llm = ChatBedrockConverse(
    region_name=os.environ["AWS_REGION"],
    aws_access_key_id=os.environ["AWS_ACCESS_KEY_ID"],
    aws_secret_access_key=os.environ["AWS_SECRET_ACCESS_KEY"],
    model_id="eu.amazon.nova-lite-v1:0",
)

llm.invoke("test", additional_model_request_fields={"inferenceConfig": {"topK": 1}})

Error Message and Stack Trace (if applicable)

/Users/user/projects/app/.venv/bin/python /Users/user/projects/app/bug_demo.py 
Traceback (most recent call last):
  File "/Users/user/projects/app/bug_demo.py", line 13, in <module>
    llm.invoke("test", additional_model_request_fields={"inferenceConfig": {"topK": 1}})
    ~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_core/language_models/chat_models.py", line 402, in invoke
    self.generate_prompt(
    ~~~~~~~~~~~~~~~~~~~~^
        [self._convert_input(input)],
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    ...<6 lines>...
        **kwargs,
        ^^^^^^^^^
    ).generations[0][0],
    ^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_core/language_models/chat_models.py", line 1121, in generate_prompt
    return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
           ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_core/language_models/chat_models.py", line 931, in generate
    self._generate_with_cache(
    ~~~~~~~~~~~~~~~~~~~~~~~~~^
        m,
        ^^
    ...<2 lines>...
        **kwargs,
        ^^^^^^^^^
    )
    ^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_core/language_models/chat_models.py", line 1233, in _generate_with_cache
    result = self._generate(
        messages, stop=stop, run_manager=run_manager, **kwargs
    )
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_aws/chat_models/bedrock_converse.py", line 1086, in _generate
    _handle_bedrock_error(e)
    ~~~~~~~~~~~~~~~~~~~~~^^^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_aws/chat_models/bedrock_converse.py", line 1621, in _handle_bedrock_error
    raise error
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/langchain_aws/chat_models/bedrock_converse.py", line 1082, in _generate
    response = self.client.converse(
        messages=bedrock_messages, system=system, **params
    )
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/botocore/client.py", line 602, in _api_call
    return self._make_api_call(operation_name, kwargs)
           ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/botocore/context.py", line 123, in wrapper
    return func(*args, **kwargs)
  File "/Users/user/projects/app/.venv/lib/python3.13/site-packages/botocore/client.py", line 1078, in _make_api_call
    raise error_class(parsed_response, operation_name)
botocore.errorfactory.ValidationException: An error occurred (ValidationException) when calling the Converse operation: The model returned the following errors: Malformed input request: #: extraneous key [inference_config] is not permitted, please reformat your input and try again.

Process finished with exit code 1

Description

Langchain turns valid arguments that complies with expected AWS request schema to invalid by forcing all nested keys to follow camel_case notation.

The code example worked correctly in previous versions. I believe this error was introduced in #829