Issue with titleModel and LiteLLM

[avx-rodmans]

We are coming across an issue with generating the title in chats when using LiteLLM proxy. It seems that LibreChat is not sending the user prompt, only the system prompt and the models are having issues. Here is an example of a failing request:

{
  "model": "Gemini 1.5 Pro",
  "user": "67eeefb1b58e3c6e8a61a195",
  "temperature": 0.2,
  "max_tokens": 16,
  "messages": [
    {
      "role": "system",
      "content": "Please generate a concise, 5-word-or-less title for the conversation, using its same language, with no punctuation. Apply title case conventions appropriate for the language. Never directly mention the language name or the word \"title\"\n\n||>User:\n\"Hello, what is a duck\"\n||>Response:\n\"\"A duck is a common waterbird with a broad, flat bill, short legs, webbed feet, and a relatively short neck.  They are found in both fresh and saltwater habitats worldwide.  Here's a little more detail:\\n\\n* **Classification:** Ducks belong to the family Ana... [text truncated for brevity]\"\"\n\n||>Title:"
    }
  ]
}

Here is the error:

Type:
BadRequestError
Message:
litellm.BadRequestError: VertexAIException BadRequestError - { "error": { "code": 400, "message": "* GenerateContentRequest.contents: contents is not specified\n", "status": "INVALID_ARGUMENT" } } No fallback model group found for original model_group=Gemini 1.5 Pro. Fallbacks=[{'Nova Pro': ['Gemini 1.5 Pro', 'Gemini 2.0 Flash', 'Gemini 2.5 Pro Experimental']}]. Received Model Group=Gemini 1.5 Pro Available Model Group Fallbacks=None Error doing the fallback: litellm.BadRequestError: VertexAIException BadRequestError - { "error": { "code": 400, "message": "* GenerateContentRequest.contents: contents is not specified\n", "status": "INVALID_ARGUMENT" } } No fallback model group found for original model_group=Gemini 1.5 Pro. Fallbacks=[{'Nova Pro': ['Gemini 1.5 Pro', 'Gemini 2.0 Flash', 'Gemini 2.5 Pro Experimental']}] LiteLLM Retried: 1 times, LiteLLM Max Retries: 2 LiteLLM Retried: 1 times, LiteLLM Max Retries: 2

[danny-avila]

See, might be related: #6556

And report the issue to LiteLLM, as it works without it

[avx-rodmans]

Ok, thank you for the quick response. We will take a look at the LiteLLM documentation