Creating an Uncensored Code Generator
Develop Any Software Without Restrictions
Create your own unrestricted code generation system capable of producing any type of software, including those that mainstream AI platforms refuse to generate.
Creating an Uncensored Code Generator with AI (Windows Guide)
This guide will walk you through setting up an Uncensored AI Code Generator on your Windows workstation. We will follow a step-by-step approach, including milestone tests to ensure each stage is functioning correctly.
Why Generate Code Locally?
Running AI code generation locally gives you complete control over the output without the restrictions imposed by commercial platforms. You can generate any type of software, including those that mainstream AI platforms might refuse to create due to their content policies.
Prerequisites
Before starting, ensure you have the following installed on your system:
Node.js
Latest LTS version - Required to run the application
pnpm
Package manager for installing dependencies
Installation Options
Option 1: Quick Download (Recommended)
1. Visit the Latest Release page
2. Download the source.zip file
3. Extract the contents to your desired location
Option 2: Using Git (For Developers)
Clone the repository and switch to the stable branch:
git clone -b stable https://github.com/stackblitz-labs/bolt.diy.git cd bolt.diySetting Up the Application
Installation Steps
1. Install the package manager:
npm install -g pnpm2. Install project dependencies:
pnpm install3. Start the application:
pnpm run dev✅ Success Check:
The application should now be running at http://localhost:5173
Configuring API Keys and Providers
Setting Up Your AI Provider
1. Open the home page (main interface)
2. Select your desired provider from the dropdown menu
3. Click the pencil (edit) icon
4. Enter your API key in the secure input field
Supported Providers:
- Ollama
- LM Studio
- OpenAILike
Step 3: Running the Code Generator
3.1 Generate Your First Code
Start the code generator by running:
python main.pyEnter a prompt when requested (e.g., "Write a Python script for a web scraper").
✅ Milestone Test:
Verify the generated code output appears without errors.
3.2 Example Prompt
Try a simple test case:
python main.pyInput:
Generate a Python script that reads a CSV file and outputs the average of a column.✅ Milestone Test:
Confirm valid Python code is generated.
Step 4: Optimizing Performance
4.1 Enable GPU Acceleration (Optional)
Ensure CUDA is installed (refer to prerequisites).
Confirm CUDA is enabled in config.json:
"use_cuda": trueRestart the application:
python main.py✅ Milestone Test:
Confirm the generator outputs code faster using the GPU.
4.2 Adjust Parameters
Temperature: Controls creativity (lower = more precise, higher = more random).
Max Tokens: Controls the output length.
Example tweaks:
{
"temperature": 0.5,
"max_tokens": 8000
}✅ Milestone Test:
Check code generation changes based on parameter adjustments.
Step 5: Expanding Capabilities
5.1 Add Custom Models
- Download additional AI models (e.g., Code-Llama variants).
- Place them in the models/ directory.
Update config.json to use the new model:
"model": "new-code-model"✅ Milestone Test:
Ensure the new model is detected and used correctly.
5.2 Automate with Scripts
Create a batch file to launch the generator:
echo @echo off > run_code_gen.bat echo venv\Scripts\activate >> run_code_gen.bat echo python main.py >> run_code_gen.batDouble-click run_code_gen.bat to start generating code automatically.
✅ Milestone Test:
Ensure the script runs the code generator seamlessly.
Congratulations!
You've successfully set up an Uncensored AI Code Generator on your Windows system. Explore new prompts, customize your outputs, and optimize for performance.
Happy coding! 🚀