Save responses from multiple LLMs

[tubamos]

Notes

This is basically an extention to issue 149 which is tasked to save the conversation from the user when they interact with one LLM.

• Create option to save / export the conversation as formated txt #149

In the current one we are seeking to extend the saving functionality to the responses of all the LLMs that are activated by the user via the selection checkbox.

Dependencies

• Query and display multiple LLM responses simultaneously - frontend #204

• Query and display multiple LLM responses simultaneously - backend #208

• New functions architecture - Refactoring - Query and display multiple LLM responses simultaneously - backend #266

Domain

app backend
app frontend

Description

Add functionality to the existing save button in the app to export responses from multiple LLMs simultaneously. When more than one LLMs have been selected in the dropdown and responses have been generated by all of them, by pressing the save button on the top bar, all the responses get to be saved / exported to a single txt file or several txt files (one per LLM interaction) whichever option (one or serveral) is easier to implement.

User Story

• As a user,
• I want the save button to export all selected LLM responses into a text file,
• so that I can have a convenient way to review and compare the answers.

Acceptance Criteria

• Users can select multiple LLMs in the dropdown whose responses will be saved when the save button is pressed.
• The save button allows users to export responses from multiple selected LLMs into a single or several text file(s).
• The exported text file contains clearly labeled sections for each LLM’s response, indicating which LLM provided each answer to make the comparison of responses from different LLMs easier. The use of various delimiters, white space etc. makes the legibility of long text files easier for the user.

Definition of Done

• The feature has been fully implemented.
• The feature has been manually tested and works as expected without critical bugs.
• The feature code is documented with clear explanations of its functionality and usage.
• The feature code has been reviewed and approved by at least one team member.
• The feature branches have been merged into the main branch and closed.
• The feature utility, function and usage have been documented in the respective project wiki on github.