AzureChatOpenAI - Not currently supporting o3-mini?

[aldensiol]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

def get_model(
    model_type: Literal["mini_llm_model", "standard_llm_model", "smart_llm_model"]
):
    if model_type == "smart_llm_model":
        return AzureChatOpenAI(
            api_key=API_KEY,
            azure_endpoint=ENDPOINT,
            api_version="2025-01-31",
            deployment_name=model_config.get(model_type, "o3-mini"),
        )
    return AzureChatOpenAI(
        api_key=API_KEY,
        azure_endpoint=ENDPOINT,
        api_version=API_VERSION,
        deployment_name=model_config.get(model_type, "gpt-4o-mini-2024-07-18"),
        rate_limiter=InMemoryRateLimiter(
            requests_per_second=rate_config.get("requests_per_second", 5),
            check_every_n_seconds=rate_config.get("check_every_n_seconds", 1),
            max_bucket_size=rate_config.get("max_bucket_size", 1),
        ),
        **{
            "temperature": model_config.get("temperature", 0.1),
            "max_tokens": model_config.get("max_tokens", 256),
            "top_p": model_config.get("top_p", 1),
            "frequency_penalty": model_config.get("frequency_penalty", 0),
            "presence_penalty": model_config.get("presence_penalty", 0),
        }
    )

gpt_o3_mini = get_model(model_type="smart_llm_model")

Description

Hey friends, after trying to invoke gpt_o3_mini via AzureChatOpenAI, I am getting a 400 error:
"code: unsupported_parameter; message: Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.", 'type': 'invalid_request_error', 'param': 'max_tokens', 'code': '-4003'}}

I've already tried updating the langchain-openai library. Is this a deprecated library issue, and how can i possibly override the implementation to fix it?

System Info

Package Information

langchain_core: 0.3.37
langchain: 0.3.15
langchain_community: 0.3.15
langsmith: 0.3.8
langchain_openai: 0.3.6
langchain_text_splitters: 0.3.5
langgraph_sdk: 0.1.51