# OpenAI

The NLX OpenAI Chat channel acts as a compatibility layer (proxy) that allows standard LLM client libraries to communicate directly with your NLX AI apps.

This allows you to swap out "pure" LLM calls in your existing applications (like those using LangChain, AutoGen, or the OpenAI SDK) with calls to your managed NLX app, without rewriting your integration logic.

#### Key differences

Unlike the Standard REST API, this interface mimics the OpenAI API specification strictly

1. *Authentication*: Uses `Bearer` token auth (standard for OpenAI clients)
2. *Endpoint*: Mimics `/chat/completions`
3. *Model Parameter*: Used to pass routing and session information

### Configuration

| Setting      | Value                                                                |
| ------------ | -------------------------------------------------------------------- |
| **Base URL** | `https://apps.nlx.ai/v1/{deploymentKey}/{channelKey}-{languageCode}` |
| **API Key**  | Your NLX API Key (passed as Bearer token)                            |

{% hint style="info" %}
You must append the language code (e.g., `-en-US`) to the end of your channel key in the Base URL.
{% endhint %}

#### The `model` parameter

In standard OpenAI calls, `model` specifies the engine (e.g., `gpt-4`). In NLX OpenAI Connect, the `model` parameter is used to maintain conversation state.

Format: `nlx:{conversationId}`

* `nlx:61d4735d-d67b-4496-8e41-3094b7c5e1d5`: Resume or continue conversation `61d4735d-d67b-4496-8e41-3094b7c5e1d5`
* `nlx`: (Omit ID) Start a brand new session. The API will generate a UUID for you

### **Integration examples**

**OpenAI Node.js SDK**

You can use the official OpenAI library by simply changing the `baseURL` and `apiKey`.

```
import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'YOUR_NLX_API_KEY', // Use your NLX key here
  baseURL: 'https://apps.nlx.ai/v1/xxxx/xxxx-en-US'
});

async function main() {
  const stream = await client.chat.completions.create({
    model: 'nlx:61d4735d-d67b-4496-8e41-3094b7c5e1d5', // Pass session ID here
    messages: [{ role: 'user', content: 'Hello!' }],
    stream: true,
  });

  for await (const chunk of stream) {
    process.stdout.write(chunk.choices[0]?.delta?.content || '');
  }
}

main();
```

**LangChain (Python)**

Easily plug an NLX App into a LangChain workflow.

```
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    openai_api_key="YOUR_NLX_API_KEY",
    openai_api_base="https://apps.nlx.ai/v1/xxxx/xxxx-en-US",
    model_name="nlx:61d4735d-d67b-4496-8e41-3094b7c5e1d5"
)

response = llm.invoke("How do I return a product?")
print(response.content)
```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.nlx.ai/platform/developers/conversation-api/openai.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.