[Bug] LM Studio Model

[bukit-kronik]

Describe the bug

Use "Default model" placeholder for LM Studio so I can save settings.

[johnfunmula]

Can you try 0.3.2? It should solve your problem.

By the way, this screenshot, if you open LM studio and browse to http://localhost:1234/v1/models, will it be empty?

[bukit-kronik]

Will try ASAP. Btw, LM Studio doesn't have a strict model naming, that's why I ask just a "default model", should work whatever model loaded.

[bukit-kronik]

@johnfunmula It works.. Thank you.

but it doesn't work when chat with the model, unfortunately.

2025-02-18 16:41:37 [DEBUG] 
Received request: POST to /v1/chat/completions with body  {
  "model": "qwen2.5-7b-celestialharmony-1m-i1",
  "temperature": 1,
  "top_p": 1,
  "frequency_penalty": 0,
  "presence_penalty": 0,
  "n": 1,
  "stream": true,
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "echo",
        "description": "A simple echo tool to verify if the MCP server is ... <Truncated in logs> ...racteristic response containing the input message.",
        "parameters": {
          "type": "object",
          "properties": {
            "message": {
              "type": "string",
              "description": "Message to be echoed back"
            }
          },
          "required": [
            "message"
          ],
          "additionalProperties": false
        }
      }
    }
  ],
  "stream_options": {
    "include_usage": true
  },
  "messages": [
    {
      "role": "system",
      "content": "\n<Dive_System_Thinking_Protocol>\n  I am an AI Assi... <Truncated in logs> ..._Specific_Rules>\n</Dive_System_Thinking_Protocol>\n"
    },
    {
      "role": "user",
      "content": "hello"
    },
    {
      "role": "assistant",
      "content": "."
    },
    {
      "role": "user",
      "content": "hello?"
    },
    {
      "role": "assistant",
      "content": "."
    },
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "hello"
        }
      ]
    }
  ]
}
2025-02-18 16:41:37  [INFO] 
[LM STUDIO SERVER] Running chat completion on conversation with 6 messages.

[johnfunmula]

@bukit-kronik Thank you for report. We'll try to check how these models behave in the program.