Enhancement: Pure code output formatting for Ollama models

[clstaudt]

What features would you like to see added?

I have added an Ollama endpoint and am testing it with the orca-mini:13b model.

I would like to see the code in the output properly formatted (font, indentation, syntax highlighting)

More details

    - name: "Ollama"
      apiKey: "ollama"
      # use 'host.docker.internal' instead of localhost if running LibreChat in a docker container
      baseURL: "http://host.docker.internal:11434/v1/chat/completions" 
      models:
        default: [
          "orca-mini:13b",
          "dolphin-mixtral:latest",
          "llama3:latest",
          "starcoder2:15b",
          ]
        fetch: false # fetching list of models is not supported
      titleConvo: true
      titleModel: "orca-mini:13b"
      summarize: false
      summaryModel: "orca-mini:13b"
      forcePrompt: false
      modelDisplayLabel: "Ollama"

Not entirely sure if this is an issue of the model or of LibreChat.

Which components are impacted by your request?

General

Pictures

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[clstaudt]

In fact this is already working, if the model outputs text first and code second (assuming in tripe backticks?). Would be nice if pure code output would be formatted properly as well.

[danny-avila]

this is all on the model training. You can add a custom instruction for this via preset:
"Please always output multi-line code in markdown format with backticks"