Suggested model to run locally in LM Studio with this course?

Short Course Q&A Building AI Browser Agents
,
1

I’m attempting to run this course locally interfacing with LM Studio server. It was easy enough to setup and run, I just modified the line in the 2nd code cell where it set the value of “client” to:

client = OpenAI(base_url=“http://localhost:1234/v1”) #Had to break the link to post this

The issue is that I tried using the openai/gpt-oss-20b model and it wasn’t able to extract any course data.

I know it’s a different model than what the course is using, but can anyone provide an explanation on why this model wouldn’t be able to run through the same steps as the OpenAI model used in the original notebook?

Here’s the log output from the developer Logs in LM Studio, maybe this could help:

2025-09-26 00:29:14  [INFO]
 [openai/gpt-oss-20b] Generated prediction:  {
  "id": "chatcmpl-17dehcdg0fymapo1ft1ozn",
  "object": "chat.completion",
  "created": 1758860929,
  "model": "openai/gpt-oss-20b",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "{\"courses\":[] }",
        "reasoning": "",
        "tool_calls": []
      },
      "logprobs": null,
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 22553,
    "completion_tokens": 8,
    "total_tokens": 22561
  },
  "stats": {},
  "system_fingerprint": "openai/gpt-oss-20b"
}