Bolt.new: AI-powered full-stack web development
LLM-powered browser IDE with integrated WebContainers runtime.
Learn more about bolt.new
Bolt.new is a web application that integrates large language models with StackBlitz's WebContainers runtime to provide an interactive development environment in the browser. The system gives AI agents control over the filesystem, package manager, Node.js server, and terminal, allowing it to handle the complete application lifecycle from scaffolding through deployment. It supports most JavaScript frameworks and libraries that run on StackBlitz, including Vite, Next.js, Astro, and others. The tool targets developers, designers, and product managers who want to prototype or build applications without managing local development infrastructure.
Browser-Based Runtime Environment
StackBlitz WebContainers execute Node.js, npm packages, and development servers entirely in-browser without backend infrastructure. Eliminates local setup while enabling full-stack development including third-party API integration and live server execution.
AI Environment Control
The language model directly manipulates filesystem, package manager, and terminal to observe execution results and iterate on code. Goes beyond code generation by closing the feedback loop between implementation and runtime behavior.
Deployment from Chat
Production deployments with shareable URLs launch directly from the conversational interface. Integrates the complete development lifecycle from scaffolding to deployment without context switching or external tooling.
// Prompt Bolt to create a simple React app
"Create a React todo list app with add and delete functionality"
// Bolt will:
// - Scaffold the project structure
// - Install dependencies via npm
// - Start the dev server
// - Show live preview in browserSee how people are using bolt.new
Related Repositories
Discover similar tools and frameworks used by developers
YOLOX
PyTorch anchor-free object detector with scalable model variants.
video2x
Video2X enhances video quality using machine learning algorithms for upscaling, frame interpolation, and restoration with multiple backend support.
segment-anything
Transformer-based promptable segmentation with zero-shot generalization.
ollama
Go-based CLI for local LLM inference and management.
stable-diffusion
CLIP-conditioned latent diffusion model for text-to-image synthesis.