Why Halu?
Frontend developers shouldn't have to wait for backend APIs to start building. Halu removes that bottleneck so you can move fast and stay focused on your UI.
The Problem
Backend APIs are rarely ready when you need them. Waiting on endpoints, sample data, and specs creates costly delays and blocks frontend progress.
The Solution
Halu lets you create realistic mock APIs in seconds. Define endpoints manually, generate them with AI from a plain-English description, or import your existing OpenAPI and Postman specs.
The Result
Start a mock server instantly and develop your frontend without waiting. Realistic Faker-powered responses mean your UI behaves just like it would against the real thing.
Features
Everything you need to build and manage mock APIs for development and testing.
AI-Powered Generation
Describe your API in plain English and let AI generate complete endpoint definitions with realistic mock data.
Import Collections
Import OpenAPI 3.x, Swagger 2.0, or Postman v2.0/v2.1 collections. Auto-detection picks the right parser.
Mock Server
Start a per-project mock server with slug-based routing and configurable base URL prefixes.
Per-Project Auth
Enable Bearer token authentication per project with auto-generated tokens and one-click regeneration.
Faker-Powered Data
Use ~100 Faker placeholders like {email}, {product.category}, or {lorem.paragraph} for dynamic, realistic responses.
Setup Wizard & Admin
5-step first-run wizard for auth, admin email, and AI config. Admin panel for managing AI models and settings.
Copy cURL
One-click cURL command copy with proper headers and authentication for any endpoint.
AI Chat Sidebar
Conversational AI sidebar for iterative endpoint creation. Follow-up messages modify existing endpoints.
Built with Modern Technologies
A full-stack platform powered by industry-leading tools.
Frontend
React 19, TypeScript, Vite, Tailwind CSS, shadcn/ui
Backend
Python, FastAPI, MongoDB, Pydantic v2
AI
OpenRouter API with configurable models
Auth
Supabase (Email, GitHub, Google OAuth)
Infrastructure
Docker Compose, Nginx, MongoDB 7
Data
Faker.js with ~100 placeholder mappings
How It Works
Get up and running in four simple steps.
Create a Project
Set up a new project with a name, description, and unique slug for your mock server.
Define Endpoints
Use the AI chat sidebar, import a collection, or manually configure your API endpoints with Faker placeholders.
Start the Mock Server
Launch your mock server with one click. Access endpoints via slug-based routing at /{slug}/your-path.
Integrate & Build
Point your frontend to the mock server URL. Copy cURL commands, enable auth tokens, and start building.