Getting Started
Choose Your Path
Get started fast: Sign up at scrapalot.app and use the web app - no installation needed! On your phone: Install the native Android app (download the APK). Developers: Check Architecture to understand the system
Platforms
Scrapalot runs as a web app (primary, any modern browser) and a native Android app (available now, sideload the APK). A native desktop app (Windows/macOS) is coming soon - not yet downloadable. See Apps & Platforms.
What is Scrapalot?
Scrapalot is a research intelligence platform that transforms documents into queryable knowledge:
Research-grade RAG
22 strategies + 9 orchestrators, tri-modal fusion, knowledge graphs
Natural language queries
Ask complex questions, get synthesized answers with citations
Team collaboration
Shared workspaces, notes, and conversations
Academic integration
Direct import from Google Scholar, Semantic Scholar, arXiv
Privacy first
Your research data is always yours - export anytime
How It Works
- Upload or Connect - PDFs, EPUBs, Word docs, or link Google Drive, Dropbox, academic databases
- Ask Questions - Natural language queries across all your documents
- Get Insights - AI synthesizes answers from multiple sources with exact citations
Key Capabilities
For Researchers
Literature review acceleration
Query 50+ papers simultaneously
Citation tracking
Automatic source attribution and relationship mapping
Academic connectors
Import directly from Google Scholar, Semantic Scholar, arXiv
Knowledge graphs
Visualize relationships between concepts and papers
Collaborative notes
Team synthesis and discussion
For Technical Teams
Agentic RAG system
10+ perfectly crafted strategies and orchestrators with intelligent routing
Tri-modal fusion
Dense semantic (pgvector) + sparse lexical (BM25) + graph (Neo4j)
Context expansion
70% hallucination reduction through hierarchical document intelligence
15 chunking strategies
Including Contextual Retrieval and Late Chunking
Multi-provider AI
OpenAI, Anthropic, Google, Ollama, local GGUF models
Full REST API
Programmatic access to all features
Prerequisites
Minimum Requirements:
- Node.js 18+ and npm (frontend)
- Python 3.12.8 (backend)
- PostgreSQL 14+ with pgvector extension (or SQLite for development)
Optional Components:
- Docker - Easier deployment
- Redis - Caching and session management
- Neo4j - Knowledge graph features
- GPU - For local model inference (NVIDIA CUDA or Apple Metal)
Next Steps
Quick Setup (10 minutes)
Quick Start Guide - Get running fast with minimal configuration
Learn the System
User Guide - Learn how to use Scrapalot Architecture - Understand the technical design 
Deployment - Deploy to production
Deployment Options
Cloud (the standard way to use Scrapalot)
- Managed hosting - no setup required
- All agentic RAG strategies and research features
- Academic database connectors
- Your research data is always yours - export and migrate anytime
Native apps
Day to day, use the web app in any browser, or the native Android app (available now). A native desktop app is coming soon.
Best for: Individual researchers, students, and teams who want cloud convenience without infrastructure management
Cloud Researcher Plan (Free)
- Managed hosting - no setup
- Single-user workspace
- All research features
- Academic database connectors
- Generous usage limits
Best for: Researchers who want cloud convenience without infrastructure management
Professional & Enterprise (Paid)
Team collaboration and multi-user management with deployment options:
- Professional - $19/mo managed cloud for small teams
- Enterprise - Managed cloud, plus on-premise deployment to your own infrastructure on demand, custom pricing
Best for: Research teams, institutions, organizations requiring team collaboration
Getting Help
FAQ - Common questions answered GitHub Issues - Report bugs GitHub Discussions - Ask questions 
Email - support@scrapalot.com