how to set OPENAI_API_BASE or OPENAI_PROXY?

[gaoyuan5251]

Like this module ‘OpenAIEmbeddings()‘，I have bolded these parameters, how to set the address after custom encapsulated?
Such as 'http://xxx.cn/embedding'。I found in the process of using, no matter how I set up will give me the final URL to assemble, but actually is not what I want.

import os
os.environ["OPENAI_API_TYPE"] = "azure"
os.environ["OPENAI_API_BASE"] = "https://<your-endpoint.openai.azure.com/"
os.environ["OPENAI_API_KEY"] = "your AzureOpenAI key"
os.environ["OPENAI_API_VERSION"] = "2023-05-15"
os.environ["OPENAI_PROXY"] = "http://your-corporate-proxy:8080"

[dosubot[bot]]

Answer generated by a 🤖

Answer

Hello,

Based on the context you've provided, it seems you're trying to set the "OPENAI_API_BASE" and "OPENAI_PROXY" environment variables for the OpenAIEmbeddings class in the LangChain framework.

To set these environment variables, you can do so when creating an instance of the ChatOpenAI class. The openai_api_base and openai_proxy parameters of the class constructor can be used to set these environment variables. Here is an example:

from langchain.chat_models.openai import ChatOpenAI

openai = ChatOpenAI(
    model_name="your-model-name",
    openai_api_key="your-api-key",
    openai_api_base="http://your-custom-endpoint",
    openai_proxy="http://your-proxy-url"
)

In the above code, replace "your-model-name", "your-api-key", "http://your-custom-endpoint", and "http://your-proxy-url" with your actual model name, OpenAI API key, custom endpoint, and proxy URL respectively.

These parameters are then used in the _client_params property of the ChatOpenAI class to set the OpenAI client parameters:

@property
def _client_params(self) -> Dict[str, Any]:
    """Get the parameters used for the openai client."""
    openai_creds: Dict[str, Any] = {
        "api_key": self.openai_api_key,
        "api_base": self.openai_api_base,
        "organization": self.openai_organization,
        "model": self.model_name,
    }
    if self.openai_proxy:
        import openai

        openai.proxy = {"http": self.openai_proxy, "https": self.openai_proxy}  # type: ignore[assignment]  # noqa: E501
    return {**self._default_params, **openai_creds}

This property is then used in various methods of the ChatOpenAI class to make requests to the OpenAI API.

For more information, you can refer to the following files in the LangChain repository:

• libs/langchain/langchain/embeddings/localai.py
• libs/langchain/langchain/llms/openai.py
• libs/langchain/langchain/chat_models/openai.py

Also, there are similar issues that have been solved in the LangChain repository that you might find helpful:

• Issue: OpenAIEmbeddings class about Azure Openai usage example error
• langchain AzureOpenAI Embeddings
• Ability to user Azure OpenAI

I hope this helps! If you have any other questions, feel free to ask.

Best,
Dosu

Sources

• libs/langchain/langchain/embeddings/localai.py
• libs/langchain/langchain/llms/openai.py
• libs/langchain/langchain/chat_models/openai.py
• libs/langchain/langchain/chat_models/azure_openai.py

---

This response is meant to be useful, save you time, and share context. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

[gaoyuan5251]

You don't look like a robot 😂

[haseeb-heaven]

Seriously now bots are commenting on these PR's wow 😮

[gaoyuan5251]

Mainly because the OpenAIEmbeddings module will concatenate "embeddings" by default, but our embedding api is "embedding".

[haseeb-heaven]

@dosu-bot How to set API Key to Empty when changing the openai_api_base because its optional parameter here.

[devstein]

Hey @haseeb-heaven as of now Dosu only responds to the discussion author. We're working on a way for users to "fork" the conversations without creating a new discussion.

If you have any ideas or feedback about Dosu, let me know!

[haseeb-heaven]

Okay thanks for reply.