Starting a demo from JS SDK and local running Langfuse, but cannot create init trace currently

[JamieLee0510]

I have run the Langfuse in local, not by docker.
And then try the JS SDK example like below :
(not using openai because I want to migrate to other LLM)

import { Langfuse } from "langfuse";
import OpenAI from 'openai'
import { config } from 'dotenv'
config()

const langfuse = new Langfuse({
  secretKey: process.env.LANGFUSE_SECRET_KEY,
  publicKey: process.env.LANGFUSE_PUBLIC_KEY,
  baseUrl: "http://localhost:3000"
});

const openaiClient = new OpenAI({apiKey:process.env.OPENAI_KEY})

const demoFunc = async()=>{
  const trace = langfuse.trace({
    name: "my-AI-application-endpoint",
  });
   
  // Example generation creation
  const llmConfig = {
    model: "gpt-3.5-turbo",
  }
  const messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": "Who won the world series in 2020?"},
    {"role": "assistant", "content": "The Los Angeles Dodgers won the World Series in 2020."},
    {"role": "user", "content": "Where was it played?"}
  ]
  const generation = trace.generation({
    name: "chat-completion",
    model: "gpt-3.5-turbo",
    ...llmConfig,
    input: messages,
  });
    const chatCompletion = await openaiClient.chat.completions.create({messages, ...llmConfig});
   
  generation.end({
    output: chatCompletion,
  });
}

demoFunc()

any guiding would help a lot~
Besides, if the SDK document would display the demo like below would be more helpful

[marcklingen]

How did you solve this?

[JamieLee0510]

not yet... I just think it might be more proper to discuss in discord🥹
Could you help delete this thread?

Besides, I am trying using plain JS to mock the request from SDK and Langfuse test-cases