The native AI workspace for macOS
One native app for every AI model. Interactive charts, sortable tables, threaded conversations, and agentic tools — not just another chat wrapper.
Rich Content & Embedded Media
AI-generated itineraries with inline photos and formatted details
Dynamic Tables
Sortable, structured data rendered right in your conversation
Powerful Built-in Tools
Web search, image search, MCP servers, and charts — all one click away
Interactive Charts & Diagrams
Visualize financial data with Chart.js and TradingView charts
Threaded Conversations
Branch into side threads for deeper analysis without losing context
Agentic Research
Multi-step tool calling that fetches, analyzes, and presents findings
Gallery Layout
Photo-rich responses with image grids fetched and arranged automatically
Features
More than a chat window
Charts, tables, threads, agentic tools, and system-wide access — built natively for macOS.
Rich UI, Not Just Chat
Interactive charts, TradingView financials, sortable tables, image galleries, and inline Markdown editing — responses that go far beyond plain text.
Agentic Tools & MCP
Multi-step tool calling, built-in web search, and one-click MCP server connections. Your AI researches, analyzes, and presents findings autonomously.
Threads & @Mentions
Slack-like threaded conversations keep context organized. Type @claude or @gpt mid-chat to get a one-off response from a different model.
System-Wide AI
Floating prompt panel lets you run AI completions on selected text in any app — without switching windows. Plus global hotkeys and menu bar access.
Fork & Branch Chats
Branch sub-conversations from any message to explore different directions without losing your original thread.
Prompt Library
Save your best prompts with {{variable}} placeholders, organize by category, and reuse them instantly with Cmd+/.
Every Model, One App
Ollama, Claude, OpenAI, Gemini, Grok, Groq, OpenRouter, or any OpenAI-compatible API — switch instantly without leaving the app.
Apple Silicon Optimized
Native SwiftUI built for macOS. Hardware-accelerated inference via Metal for blazing-fast local model performance.
Private by Default
Run models locally with Ollama — no telemetry, no data collection. API keys stored in the macOS Keychain. Your data stays yours.
What's New
Latest updates
Version 1.2.3 (Public Beta)
Native maps integration, videos and photos web search, interactive charts and tables.
Version 1.0 — Hello, World (Private Beta)
The first public beta release of Elvean. Local AI inference, prompt templates, and a beautiful native macOS experience.
Get Started
Ready to try Elvean?
Private AI that just works. Download, open, chat — no account required.