SillyTavern: Integrating a multimodal LLM front-end interface, AI interaction and role-playing

General Introduction

SillyTavern is a natively installed interface for advanced users that allows users to interact with a wide range of large-scale language models (LLMs) and image generation engines. It supports a variety of APIs, including KoboldAI, Horde, NovelAI, and others, and offers mobile-friendly layouts, visual novel mode, automatic translation, and more.Started in February 2023 as a fork of TavernAI 1.2.8, SillyTavern now boasts over 100 contributors after two years of independent development.

Function List

• Support for multiple LLM APIs (e.g. KoboldAI, Horde, NovelAI, etc.)
• Mobile-friendly layout
• visual novel format
• automatic translation function
• Image Generation Engine Integration (Automatic1111 &) ComfyUI API)
• Text-to-speech (TTS)
• Customizable user interface
• Support for third-party extensions

Using Help

Installation process

1. environmental preparation: Ensure that the latest version of Node.js LTS is installed.
2. Get source code: Open a terminal and run the following command to clone the code base:

   git clone https://github.com/SillyTavern/SillyTavern.git -b release
   

3. Installation of dependencies: Go to the project directory and run the following command to install the dependencies:

   cd SillyTavern
   npm install
   

4. launch an application: Run the following command to start SillyTavern:

   npm start
   

Guidelines for use

1. Introduction to the interface::
   • main interface: Contains a text input box, dialog recording area, and function buttons.
   • Setup Menu: You can configure API keys, select models, adjust the interface layout, and more.
   • character card: Used to save and manage conversation history and settings for different roles.
2. Interaction with LLM::
   • Enter your question or instruction in the text input box and click the Send button.
   • The system generates a response based on the configured model and displays it in the dialog log area.
3. Image Generation::
   • Enter an image generation command in the input box (e.g. "Generate a picture of a cat") and click Send.
   • The system calls the configured image generation engine to generate and display the image.
4. text-to-speech::
   • Enable the TTS function in the Setup menu.
   • After entering the text, click on the TTS button and the system will convert the text to speech for playback.
5. automatic translation::
   • Enable the automatic translation function in the settings menu.
   • After entering the foreign language text, the system will automatically translate and display the result.
6. Extended functionality::
   • SillyTavern supports enhancements through third-party extensions.
   • Extensions can be managed and installed in the settings menu.

common problems

• How do I update SillyTavern?
  • Go to the project directory and run the following command to update the code base:

    git pull
    npm install
    npm start
    

• How do I switch branches?
  • Run the following command to switch to the staging branch:

    git checkout staging
    npm install
    npm start
    

• How do I report a problem?
  • Submit an issue on the GitHub project page describing the issue details and steps to reproduce it.