Skip to content

saveli/simple-chat

 
 

Repository files navigation

Simple Chat

Simple Chat is a user-friendly web interface that facilitates and optimizes the integration of LLMs into online experiments. Simple Chat offers three core benefits:

  1. An embeddable chat interface that eliminates the need for custom API coding,
  2. a consistent interface across survey platforms, and
  3. streaming responses that maintain conversational flow and reduce participant dropout.

Below are diagrams illustrating the architecture and chat interface:

Architecture

Architecture

Architecture

Chat interface

Chat interface

Chat

Requirements

  • Install Docker version 24.0.5, build ced0996 or newer
  • Install Docker Compose plugin version v2.20.2-desktop.1 or newer
  • Define secrets
    • Create .env file in the backend directory by copying the env.example file:

      cp backend/env.example backend/.env

      All variables defined in settings.py must be set in the .env file. AZURE_OPENAI_BASE_URL and AZURE_OPENAI_API_KEY can be left empty if not using Azure OpenAI.

    • Create .env file in the frontend directory by copying the env.example file:

      cp frontend/env.example frontend/.env

      The default values can be used for local development.

Installation

  • Refer to the deployment documentation for first-time installation.
  • Refer to the contributing documentation to set up a development environment for modifying the code.

Usage

Refer to the documentation folder for usage instructions.

Citation

If you use Simple Chat in your research, please cite as follows:

  • BibTeX:

    @misc{schettino2025facilitatingintegrationllmsonline,
          title={Facilitating the Integration of LLMs Into Online Experiments With Simple Chat},
          author={R. Bermudez Schettino and A. Dasmeh and L. Brinkmann},
          year={2025},
          eprint={2511.19123},
          archivePrefix={arXiv},
          primaryClass={cs.HC},
          url={https://arxiv.org/abs/2511.19123},
    }
    
  • APA:

    Bermudez Schettino, R., Dasmeh, A., & Brinkmann, L. (2025). Facilitating the Integration of LLMs Into Online Experiments With Simple Chat. arXiv. https://doi.org/10.48550/arXiv.2511.19123

Related

License

CC BY 4.0

About

HTML-based interface that empowers researchers to seamlessly integrate interactions with LLMs into their online experiments.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages

  • Python 42.7%
  • JavaScript 33.4%
  • TypeScript 17.8%
  • Shell 2.3%
  • HTML 1.8%
  • Dockerfile 1.6%
  • CSS 0.4%