bolt.diy

Welcome to bolt.diy, the official open source version of Bolt.new, which allows you to choose the LLM that you use for each prompt! Currently, you can use OpenAI, Anthropic, Ollama, OpenRouter, Gemini, LMStudio, Mistral, xAI, HuggingFace, DeepSeek, Groq, Cohere, Together, Perplexity, Moonshot (Kimi), Hyperbolic, GitHub Models, Amazon Bedrock, and OpenAI-like providers - and it is easily extended to use any other model supported by the Vercel AI SDK! See the instructions below for running this locally and extending it to include more models.

Check the bolt.diy Docs for more official installation instructions and additional information.

Also this pinned post in our community has a bunch of incredible resources for running and deploying bolt.diy yourself!

We have also launched an experimental agent called the “bolt.diy Expert” that can answer common questions about bolt.diy. Find it here on the oTTomator Live Agent Studio.

bolt.diy was originally started by Cole Medin but has quickly grown into a massive community effort to build the BEST open source AI coding assistant!

Join the Community
Recent Major Additions
Features
Setup
Quick Installation
Manual Installation
Configuring API Keys and Providers
Setup Using Git (For Developers only)
Available Scripts
Contributing
Roadmap
FAQ

Join the community

Join the bolt.diy community here, in the oTTomator Think Tank!

Project management

Bolt.diy is a community effort! Still, the core team of contributors aims at organizing the project in way that allows you to understand where the current areas of focus are.

If you want to know what we are working on, what we are planning to work on, or if you want to contribute to the project, please check the project management guide to get started easily.

Recent Major Additions

✅ Completed Features

19+ AI Provider Integrations - OpenAI, Anthropic, Google, Groq, xAI, DeepSeek, Mistral, Cohere, Together, Perplexity, HuggingFace, Ollama, LM Studio, OpenRouter, Moonshot, Hyperbolic, GitHub Models, Amazon Bedrock, OpenAI-like
Electron Desktop App - Native desktop experience with full functionality
Advanced Deployment Options - Netlify, Vercel, and GitHub Pages deployment
Supabase Integration - Database management and query capabilities
Data Visualization & Analysis - Charts, graphs, and data analysis tools
MCP (Model Context Protocol) - Enhanced AI tool integration
Search Functionality - Codebase search and navigation
File Locking System - Prevents conflicts during AI code generation
Diff View - Visual representation of AI-made changes
Git Integration - Clone, import, and deployment capabilities
Expo App Creation - React Native development support
Voice Prompting - Audio input for prompts
Bulk Chat Operations - Delete multiple chats at once
Project Snapshot Restoration - Restore projects from snapshots on reload

🔄 In Progress / Planned

File Locking & Diff Improvements - Enhanced conflict prevention
Backend Agent Architecture - Move from single model calls to agent-based system
LLM Prompt Optimization - Better performance for smaller models
Project Planning Documentation - LLM-generated project plans in markdown
VSCode Integration - Git-like confirmations and workflows
Document Upload for Knowledge - Reference materials and coding style guides
Additional Provider Integrations - Azure OpenAI, Vertex AI, Granite

Features

AI-powered full-stack web development for NodeJS based applications directly in your browser.
Support for 19+ LLMs with an extensible architecture to integrate additional models.
Attach images to prompts for better contextual understanding.
Integrated terminal to view output of LLM-run commands.
Revert code to earlier versions for easier debugging and quicker changes.
Download projects as ZIP for easy portability and sync to a folder on the host.
Integration-ready Docker support for a hassle-free setup.
Deploy directly to Netlify, Vercel, or GitHub Pages.
Electron desktop app for native desktop experience.
Data visualization and analysis with integrated charts and graphs.
Git integration with clone, import, and deployment capabilities.
MCP (Model Context Protocol) support for enhanced AI tool integration.
Search functionality to search through your codebase.
File locking system to prevent conflicts during AI code generation.
Diff view to see changes made by the AI.
Supabase integration for database management and queries.
Expo app creation for React Native development.

Setup

If you’re new to installing software from GitHub, don’t worry! If you encounter any issues, feel free to submit an “issue” using the provided links or improve this documentation by forking the repository, editing the instructions, and submitting a pull request. The following instruction will help you get the stable branch up and running on your local machine in no time.

Let’s get you up and running with the stable version of Bolt.DIY!

Quick Installation

← Click here to go to the latest release version!

Download the binary for your platform (available for Windows, macOS, and Linux)
Note: For macOS, if you get the error “This app is damaged”, run:
```
xattr -cr /path/to/Bolt.app
```

Manual installation

Option 1: Node.js

Node.js is required to run the application.

Visit the Node.js Download Page
Download the “LTS” (Long Term Support) version for your operating system
Run the installer, accepting the default settings
Verify Node.js is properly installed:
- For Windows Users:
  1. Press Windows + R
  2. Type “sysdm.cpl” and press Enter
  3. Go to “Advanced” tab → “Environment Variables”
  4. Check if Node.js appears in the “Path” variable
- For Mac/Linux Users:
  1. Open Terminal
  2. Type this command:
```
echo $PATH
```
  3. Look for /usr/local/bin in the output

Running the Application

You have two options for running Bolt.DIY: directly on your machine or using Docker.

Option 1: Direct Installation (Recommended for Beginners)

Install Package Manager (pnpm):
```
npm install -g pnpm
```
Install Project Dependencies:
```
pnpm install
```
Start the Application:
```
pnpm run dev
```

Option 2: Using Docker

This option requires Docker and is great when you want an isolated environment or to mirror the production image.

Additional Prerequisite

Install Docker: Download Docker

Steps

Prepare Environment Variables

Copy the provided examples and add your provider keys:
```
cp .env.example .env
cp .env.example .env.local
```
The runtime scripts inside the container source .env and .env.local, so keep any API keys you need in one of those files.

Build an Image

# Development image (bind-mounts your local source when run)
pnpm run dockerbuild
# ≈ docker build -t bolt-ai:development -t bolt-ai:latest --target development .

# Production image (self-contained build artifacts)
pnpm run dockerbuild:prod
# ≈ docker build -t bolt-ai:production -t bolt-ai:latest --target bolt-ai-production .

Run the Container

# Development workflow with hot reload
docker compose --profile development up

# Production-style container using composed services
docker compose --profile production up

# One-off production container (exposes the app on port 5173)
docker run --rm -p 5173:5173 --env-file .env.local bolt-ai:latest

When the container starts it runs pnpm run dockerstart, which in turn executes bindings.sh to pass Cloudflare bindings through Wrangler. You can override this command in docker-compose.yaml if you need a different startup routine.

Option 3: Desktop Application (Electron)

For users who prefer a native desktop experience, bolt.diy is also available as an Electron desktop application:

Download the Desktop App:
- Visit the latest release
- Download the appropriate binary for your operating system
- For macOS: Extract and run the .dmg file
- For Windows: Run the .exe installer
- For Linux: Extract and run the AppImage or install the .deb package

Alternative: Build from Source:

# Install dependencies
pnpm install

# Build the Electron app
pnpm electron:build:dist  # For all platforms
# OR platform-specific:
pnpm electron:build:mac   # macOS
pnpm electron:build:win   # Windows
pnpm electron:build:linux # Linux

The desktop app provides the same full functionality as the web version with additional native features.

Configuring API Keys and Providers

Bolt.diy features a modern, intuitive settings interface for managing AI providers and API keys. The settings are organized into dedicated panels for easy navigation and configuration.

Accessing Provider Settings

Open Settings: Click the settings icon (⚙️) in the sidebar to access the settings panel
Navigate to Providers: Select the “Providers” tab from the settings menu
Choose Provider Type: Switch between “Cloud Providers” and “Local Providers” tabs

Cloud Providers Configuration

The Cloud Providers tab displays all cloud-based AI services in an organized card layout:

Adding API Keys

Select Provider: Browse the grid of available cloud providers (OpenAI, Anthropic, Google, etc.)
Toggle Provider: Use the switch to enable/disable each provider
Set API Key:
- Click the provider card to expand its configuration
- Click on the “API Key” field to enter edit mode
- Paste your API key and press Enter to save
- The interface shows real-time validation with green checkmarks for valid keys

Advanced Features

Bulk Toggle: Use “Enable All Cloud” to toggle all cloud providers at once
Visual Status: Green checkmarks indicate properly configured providers
Provider Icons: Each provider has a distinctive icon for easy identification
Descriptions: Helpful descriptions explain each provider’s capabilities

Local Providers Configuration

The Local Providers tab manages local AI installations and custom endpoints:

Ollama Configuration

Enable Ollama: Toggle the Ollama provider switch
Configure Endpoint: Set the API endpoint (defaults to http://127.0.0.1:11434)
Model Management:
- View all installed models with size and parameter information
- Update models to latest versions with one click
- Delete unused models
- Install new models by entering model names

Other Local Providers

LM Studio: Configure custom base URLs for LM Studio endpoints
OpenAI-like: Connect to any OpenAI-compatible API endpoint
Auto-detection: The system automatically detects environment variables for base URLs

Environment Variables vs UI Configuration

Bolt.diy supports both methods for maximum flexibility:

Environment Variables (Recommended for Production)

Set API keys and base URLs in your .env.local file:

# API Keys
OPENAI_API_KEY=your_openai_key_here
ANTHROPIC_API_KEY=your_anthropic_key_here

# Custom Base URLs
OLLAMA_BASE_URL=http://127.0.0.1:11434
LMSTUDIO_BASE_URL=http://127.0.0.1:1234

UI-Based Configuration

Real-time Updates: Changes take effect immediately
Secure Storage: API keys are stored securely in browser cookies
Visual Feedback: Clear indicators show configuration status
Easy Management: Edit, view, and manage keys through the interface

Provider-Specific Features

OpenRouter

Free Models Filter: Toggle to show only free models when browsing
Pricing Information: View input/output costs for each model
Model Search: Fuzzy search through all available models

Ollama

Model Installer: Built-in interface to install new models
Progress Tracking: Real-time download progress for model updates
Model Details: View model size, parameters, and quantization levels
Auto-refresh: Automatically detects newly installed models

Fuzzy Search: Type-ahead search across all providers and models
Keyboard Navigation: Use arrow keys and Enter to navigate quickly
Clear Search: Press Cmd+K (Mac) or Ctrl+K (Windows/Linux) to clear search

Troubleshooting

Common Issues

API Key Not Recognized: Ensure you’re using the correct API key format for each provider
Base URL Issues: Verify the endpoint URL is correct and accessible
Model Not Loading: Check that the provider is enabled and properly configured
Environment Variables Not Working: Restart the application after adding new environment variables

Status Indicators

🟢 Green Checkmark: Provider properly configured and ready to use
🔴 Red X: Configuration missing or invalid
🟡 Yellow Indicator: Provider enabled but may need additional setup
🔵 Blue Pencil: Click to edit configuration

Supported Providers Overview

Cloud Providers

OpenAI - GPT-4, GPT-3.5, and other OpenAI models
Anthropic - Claude 3.5 Sonnet, Claude 3 Opus, and other Claude models
Google (Gemini) - Gemini 1.5 Pro, Gemini 1.5 Flash, and other Gemini models
Groq - Fast inference with Llama, Mixtral, and other models
xAI - Grok models including Grok-2 and Grok-2 Vision
DeepSeek - DeepSeek Coder and other DeepSeek models
Mistral - Mixtral, Mistral 7B, and other Mistral models
Cohere - Command R, Command R+, and other Cohere models
Together AI - Various open-source models
Perplexity - Sonar models for search and reasoning
HuggingFace - Access to HuggingFace model hub
OpenRouter - Unified API for multiple model providers
Moonshot (Kimi) - Kimi AI models
Hyperbolic - High-performance model inference
GitHub Models - Models available through GitHub
Amazon Bedrock - AWS managed AI models

Local Providers

Ollama - Run open-source models locally with advanced model management
LM Studio - Local model inference with LM Studio
OpenAI-like - Connect to any OpenAI-compatible API endpoint

💡 Pro Tip: Start with OpenAI or Anthropic for the best results, then explore other providers based on your specific needs and budget considerations.

Setup Using Git (For Developers only)

This method is recommended for developers who want to:

Contribute to the project
Stay updated with the latest changes
Switch between different versions
Create custom modifications

Prerequisites

Install Git: Download Git

Initial Setup

Clone the Repository:

git clone -b stable https://github.com/stackblitz-labs/bolt.diy.git

Navigate to Project Directory:
```
cd bolt.diy
```
Install Dependencies:
```
pnpm install
```
Start the Development Server:
```
pnpm run dev
```
(OPTIONAL) Switch to the Main Branch if you want to use pre-release/testbranch:
```
git checkout main
pnpm install
pnpm run dev
```
Hint: Be aware that this can have beta-features and more likely got bugs than the stable release

Open the WebUI to test (Default: http://localhost:5173)

Beginners:

Try to use a sophisticated Provider/Model like Anthropic with Claude Sonnet 3.x Models to get best results

Explanation: The System Prompt currently implemented in bolt.diy cant cover the best performance for all providers and models out there. So it works better with some models, then other, even if the models itself are perfect for >programming

Future: Planned is a Plugin/Extentions-Library so there can be different System Prompts for different Models, which will help to get better results

Staying Updated

To get the latest changes from the repository:

Save Your Local Changes (if any):
```
git stash
```
Pull Latest Updates:
```
git pull 
```
Update Dependencies:
```
pnpm install
```
Restore Your Local Changes (if any):
```
git stash pop
```

Troubleshooting Git Setup

If you encounter issues:

Clean Installation:

# Remove node modules and lock files
rm -rf node_modules pnpm-lock.yaml

# Clear pnpm cache
pnpm store prune

# Reinstall dependencies
pnpm install

Reset Local Changes:

# Discard all local changes
git reset --hard origin/main

Remember to always commit your local changes or stash them before pulling updates to avoid conflicts.

Available Scripts

pnpm run dev: Starts the development server.
pnpm run build: Builds the project.
pnpm run start: Runs the built application locally using Wrangler Pages.
pnpm run preview: Builds and runs the production build locally.
pnpm test: Runs the test suite using Vitest.
pnpm run typecheck: Runs TypeScript type checking.
pnpm run typegen: Generates TypeScript types using Wrangler.
pnpm run deploy: Deploys the project to Cloudflare Pages.
pnpm run lint: Runs ESLint to check for code issues.
pnpm run lint:fix: Automatically fixes linting issues.
pnpm run clean: Cleans build artifacts and cache.
pnpm run prepare: Sets up husky for git hooks.
Docker Scripts:
- pnpm run dockerbuild: Builds the Docker image for development.
- pnpm run dockerbuild:prod: Builds the Docker image for production.
- pnpm run dockerrun: Runs the Docker container.
- pnpm run dockerstart: Starts the Docker container with proper bindings.
Electron Scripts:
- pnpm electron:build:deps: Builds Electron main and preload scripts.
- pnpm electron:build:main: Builds the Electron main process.
- pnpm electron:build:preload: Builds the Electron preload script.
- pnpm electron:build:renderer: Builds the Electron renderer.
- pnpm electron:build:unpack: Creates an unpacked Electron build.
- pnpm electron:build:mac: Builds for macOS.
- pnpm electron:build:win: Builds for Windows.
- pnpm electron:build:linux: Builds for Linux.
- pnpm electron:build:dist: Builds for all platforms.

Contributing

We welcome contributions! Check out our Contributing Guide to get started.

Roadmap

Explore upcoming features and priorities on our Roadmap.

FAQ

For answers to common questions, issues, and to see a list of recommended models, visit our FAQ Page.

Licensing

Who needs a commercial WebContainer API license?

bolt.diy source code is distributed as MIT, but it uses WebContainers API that requires licensing for production usage in a commercial, for-profit setting. (Prototypes or POCs do not require a commercial license.) If you’re using the API to meet the needs of your customers, prospective customers, and/or employees, you need a license to ensure compliance with our Terms of Service. Usage of the API in violation of these terms may result in your access being revoked.

Test commit to trigger Security Analysis workflow