Skip to content

LLM evaluation and testing tool compatible with OpenAI API

License

Notifications You must be signed in to change notification settings

yomorun/llm-alchemist

 
 

Repository files navigation

中文 README

LLM Alchemist

LLM Alchemist is an open-source tool designed to evaluate Large Language Models (LLMs) through prompt testing. With simple configuration, you can batch evaluate prompts and store all configurations securely in your local IndexedDB. The project is built using modern web technologies like React, TypeScript, Vite, and the OpenAI JS API, ensuring high performance and easy deployment.

Features

  • Batch Prompt Evaluation: Efficiently test multiple prompts with just a few clicks.
  • Secure Local Storage: All configurations are securely stored in the browser's IndexedDB, ensuring data privacy.
  • Modern Tech Stack: Built with Vite, React, and TypeScript, offering a fast and reliable development experience.
  • OpenAI JS API Integration: Seamless interaction with the OpenAI API to obtain model responses.
  • Vercel Deployment: One-click deployment to Vercel, making it easy to bring your tool online.

Getting Started

Prerequisites

  • Node.js (version 14.x or later)
  • pnpm (recommended for package management, version 7.x or later)

Installation

  1. Clone the repository:
git clone https://github.com/houjiazong/LLM-Alchemist.git
cd LLM-Alchemist
  1. Install dependencies:
pnpm install

Running the Project

  1. Copy the development proxy configuration file:
cp example.dev.proxy.config.js dev.proxy.config.js
  1. Start the development server:
pnpm dev

Building for Production

pnpm build

The output will be located in the dist folder, ready for deployment.

Deployment

You can deploy this project directly to Vercel by clicking the button below:

Deploy to Vercel

Handling CORS issues in production environment

  • After the project is deployed, you may encounter cross-origin resource sharing (CORS) issues, especially when making API requests to other domains. To solve this problem, make sure that the server corresponding to your BaseURL is configured to allow requests from the domain you deployed.

  • You can also customize the proxy service by setting the environment variable VITE_PROXY_URL, for example:

VITE_PROXY_URL=https://your-proxy-url.com?target=

When the program is running, the original address of the request will be appended to the target parameter.

License

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

About

LLM evaluation and testing tool compatible with OpenAI API

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • TypeScript 95.6%
  • JavaScript 2.6%
  • CSS 1.5%
  • Other 0.3%