Why not extend observeOpenAI to support message-level user id?

[derrickstaten]

Example from these docs:

const res = await observeOpenAI(new OpenAI(), {
  generationName: "Traced generation",
  metadata: { someMetadataKey: "someValue" },
  sessionId: "session-id",
  userId: "user-id",
  tags: ["tag1", "tag2"],
  version: "0.0.1",
  release: "beta",
}).chat.completions.create({
  messages: [{ role: "system", content: "Tell me a story about a dog." }],
  model: "gpt-3.5-turbo",
  max_tokens: 300,
});

Having userId at the config level isn't helpful, as I am building a multi-tenant app, meaning many users. It needs to be in the chat.completions.create({}) call, alongside messages, model, and max_tokens.

Steps to reproduce

Here's my code:

'use server';

import OpenAI from 'openai';
import { observeOpenAI } from 'langfuse';

const openai = observeOpenAI(
  new OpenAI({
    apiKey: process.env.OPENAI_API_KEY,
  })
);

export async function invokeChatGPT(
  prompt: string,
  userId: string
): Promise<string> {
  try {
    const completion = await openai.chat.completions.create({
      model: 'gpt-4-turbo',
      messages: [
        {
          role: 'system',
          content: 'You are a helpful assistant.',
        },
        { role: 'user', content: prompt },
      ],
      user: userId,
    });

    return completion.choices[0].message.content || 'No response from ChatGPT';
  } catch (error) {
    console.error('Error invoking ChatGPT:', error);
    throw error;
  }
}

I was hoping that user: userId would work, as it's supported by OpenAI.

Expected Results

My traces should have user set in the UI found at https://cloud.langfuse.com/project/PROJECT_ID_HERE/traces

Actual results

My traces do not have user set in the UI.

[derrickstaten]

Whoops I guess I can just do this (persisting the new OpenAI client):

'use server';

import OpenAI from 'openai';
import { observeOpenAI } from 'langfuse';

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
});

export async function invokeChatGPT(
  prompt: string,
  userId: string
): Promise<string> {
  try {
    const openai = observeOpenAI(client, { userId: userId });
    const completion = await openai.chat.completions.create({
      model: 'gpt-4-turbo',
      messages: [
        {
          role: 'system',
          content: 'You are a helpful assistant.',
        },
        { role: 'user', content: prompt },
      ],
    });

    return completion.choices[0].message.content || 'No response from ChatGPT';
  } catch (error) {
    console.error('Error invoking ChatGPT:', error);
    throw error;
  }
}