Skip to content

Simple GUI to make LLMs code for humans. Finally the time has come!

Notifications You must be signed in to change notification settings

VedAstro/LLMCodes

Repository files navigation

LLMCodes

LLMCodes is a simple graphical user interface (GUI) application designed to facilitate code generation using Large Language Models (LLMs). This project allows users to easily switch between multiple LLMs, providing a versatile coding assistant experience.

Features

  • Multiple LLM Support: Easily switch between different LLMs like GPT-4, Phi3, MistralNemo, MetaLlama, and Cohere.
  • Local Execution: Run the application locally with unlimited timeout HTTP calls, offering more flexibility and control than online alternatives.
  • Past Prompts: Load and reuse previous prompts from your chat history.
  • Simple User Interface: A straightforward and user-friendly interface for a smooth experience.

image image image

Installation

  1. Clone the Repository:
Copy code
git clone https://github.com/your-username/LLMCodes.git
cd LLMCodes
  1. Set Up Configuration:

Create a secrets.json file in the root directory with your API keys and endpoints. Use the following structure:

Copy code
{
  "GPT4oEndpoint": "your-gpt4o-endpoint",
  "GPT4oApiKey": "your-gpt4o-api-key",
  "Phi3medium128kinstructEndpoint": "your-phi3medium128kinstruct-endpoint",
  "Phi3medium128kinstructApiKey": "your-phi3medium128kinstruct-api-key",
  "MistralNemo128kEndpoint": "your-mistralnemo128k-endpoint",
  "MistralNemo128kApiKey": "your-mistralnemo128k-api-key",
  "MetaLlama31405BEndpoint": "your-metallama31405b-endpoint",
  "MetaLlama31405BApiKey": "your-metallama31405b-api-key",
  "CohereCommandRPlusEndpoint": "your-coherecommandrplus-endpoint",
  "CohereCommandRPlusApiKey": "your-coherecommandrplus-api-key"
}

Build and Run:

Open the solution in Visual Studio or your preferred .NET IDE. Build the project and run the application.

Usage

Selecting an LLM:

Use the dropdown menu to select your preferred LLM from the available options. Entering a Prompt:

Type your code-related question or prompt in the input box and click "Send". Viewing Responses:

The LLM's response will appear in the output box. Managing Chat History:

Previous user prompts can be selected from the dropdown to reuse or edit. Customizing Settings:

The settings in secrets.json can be adjusted to update API keys and endpoints.

Contributing

Contributions are welcome! Please fork the repository and submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for more details.

About

Simple GUI to make LLMs code for humans. Finally the time has come!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages