Sorana
Visual AI Workspace
Inspiration
We were frustrated by the limitations of traditional file managers. For decades, we've been forced to organize digital lives into rigid, list-based hierarchies and nested folders that hide information rather than reveal it. Our brains don't work in lists; they work through associations and spatial relationships. We wanted to build a system that reflects thisβmoving away from "where did I save that file?" to "what is this project about?" by using semantic and visual grouping to reveal the hidden structure of our data.
What is Sorana?
The first AI workspace that actually remembers you. Stop attaching files to amnesiac chat boxes. Sorana solves the "blank slate" problem by combining a 2D spatial canvas with a powerful 4-tier memory system. Find files by semantic topic, chat with your documents, and let the built-in MCP server automate your tasks (create, move, edit). Because it understands your active workspace, Sorana proactively greets you with project context, extracting facts, skills, and preferences to build a long-term profile. Run models locally so nothing leaves your machine, or connect to cloud AI.
Demonstration: Watch Sorana semantically organize files and engage with deep project memory.
π‘ What Sorana Solves
π§ The "Amnesia" Problem
π Chatting is disconnected from doing
βοΈ Cloud Privacy Risks
πΈ Token Bloat
ποΈ Scanned PDFs are blind to AI
π "Where did I save that file?"
π How Sorana Compares
AI Memory
Sorana's AI remembers who you are, what you prefer, and what you've discussed β making every answer more accurate while using dramatically fewer tokens. Perfect for local and private inference.
Memory That Learns Over Time
How It Works
Key Benefits
MCP Server & Integrations
Stop context switching. Traditional AI makes you copy-paste between windows. Through the built-in MCP (Model Context Protocol) server, your AI isn't just a chatbotβit's an operator. Give your AI agents direct, conversational control over your local files, web research, and external services like Google Drive, GitHub, or custom databases.
- Open MCP Manager to configure and enable servers
- Connect external MCP servers if needed (Google Drive, GitHub, databases, etc.)
- Create an agent in the workspace
- Right-click on the agent title and select "Chat"
- Interact directly with files, folders, web content, and external services
MCP Manager
File Operations
Web Content Tools
Gmail MCP Server
Document Processing & OCR
Problem: Scanned PDFs and images are invisible to AI β no text content to search or chat with.
Solution: OCR extracts text from scanned documents, images, and PDFs. Once processed, documents become searchable and indexed via adaptive retrieval (semantic search β keywords β full-text fallback) so AI finds precise answers from your documents. Process documents via the context menu ('Document Overview' and 'Process Documents') β particularly valuable for PDFs and image-based documents.
OCR Capabilities
Encoding Support Notes
AI Backends & Models
AI Model Performance
AI Integration
Model Manager Split View
Hardware Requirements
Recommendation
Installation & Setup
Basic installation steps:
- Download: Obtain
Sorana.exe. - Execute: Run the executable file.
- Configure: In Model Manager, select your AI model.
- Run: In main window click Organize Folder.
System Requirements
| Component | Minimum |
|---|---|
| OS | Windows 11 (64-bit) |
| π§ AI Support | Built-in models or on-prem/cloud AI services |
| πΎ RAM | ≥ 4 GB |
| π½ Storage | Minimum 2 GB (app + model) + .sorana/ data folder |
| β‘ Rights | Standard user |
π Claim Your Workspace
Stop renting your AI memory. Start owning it. Download Sorana today, run it fully offline, and build a digital workspace that actually learns how you work.
No installation required. Unzip and run. Your data stays on your machine.
$ md5sum.exe Sorana.exe v1.0.36 e04ccb2a9880f8fd31ee70bbf57ed80b


