# Vehla

> A native macOS command center with on-device AI, window management, clipboard intelligence, dictation, and more — replacing a stack of single-purpose utilities behind one hotkey.

Vehla is a native macOS command center built in Swift for Apple Silicon, designed to replace a collection of single-purpose utilities — launchers, clipboard managers, window managers, dictation apps, shelves, and notes tools — with a single interface invoked by one hotkey. It is developed by Kaila Consulting Inc. and is available as a free-forever tier with a one-time Pro upgrade.

## What It Is

Vehla sits at the operating-system layer of macOS, surfacing as a command palette (⌘Space) that can launch apps, manipulate text, search clipboard history, dictate, move windows, hold files temporarily, run AI actions, and open workspaces — all without leaving the current app. It runs natively on Apple Silicon using macOS 15.0 or later and does not require an internet connection for its core features when a local model is downloaded.

## The System: Fourteen Surfaces in One

Vehla's homepage describes the problem it solves as "fourteen separate purchases" — the stack of focused utilities that most Mac power users accumulate. Vehla consolidates these into named subsystems:

- **QuickGlass** — AI actions (summarize, rewrite, explain, reply) that appear beside selected text with no context switching
- **Shelf** — temporary working memory for files, images, folders, and URLs dropped from anywhere on the desktop
- **Flow** — on-device, streaming, system-wide voice dictation that types directly into the focused app
- **SnapWheel** — a radial gesture menu around the cursor for window snapping, AI, clipboard, and shelf actions
- **Island** — a persistent system surface at the edge of macOS for controls, status, and shortcuts
- **Clipboard** — searchable history across text, code, URLs, images, emails, and commands
- **Workspaces** — Notes, Kanban, Canvas, Draw.io, and a local Knowledge/Recall index

## Local-First AI Architecture

Vehla is built around a local-first AI model. It ships an on-device MLX runtime, supports Ollama and LM Studio, and integrates with Lekh AI Pro. Users who want cloud models bring their own API keys for OpenAI, Anthropic, Google Gemini, DeepSeek, or OpenRouter; the homepage states that requests go directly from the Mac to the provider using the user's own key, stored in macOS Keychain. Vehla describes itself as sitting "above the model layer so the workflow stays the same" regardless of whether a local or cloud model is active.

## Extensions and MCP Integration

Vehla supports a native extension system with TypeScript/JavaScript and Swift SDKs, published under MIT license at the `ibuhs/vehla-extensions` GitHub repository. Extensions can add commands, native SwiftUI/AppKit workspaces, and Dock widgets. The platform also supports MCP servers, custom AI actions, browser integration, and Finder actions. The extension catalog is browsable and installable directly from Vehla's Settings → Store panel, with SHA-256 checksum verification and optional Ed25519 publisher signing for catalog packages.

## How It Compares to Raycast and Alfred

Vehla's homepage includes a self-published feature matrix comparing it to Raycast, Alfred, and SuperCmd across palette, AI, clipboard, windows, notch/shelf, menu bar, and notes/canvas dimensions. According to that matrix, Vehla is the only entry with built-in support across all seven categories. The homepage explicitly notes that "Raycast has the largest extension catalog and hosted AI on a subscription; Alfred has the deepest workflow automation" and that Vehla's differentiators are the on-device model, the surfaces beyond the palette, and a one-time price model.

## Current Status and Platform Requirements

Vehla requires macOS 15.0 or later on Apple Silicon and does not run on Intel Macs. The extensions repository was created in July 2026 and last updated in September 2026, indicating active development. The homepage notes a 7-day Pro trial with no credit card required and a refund policy within 7 days of purchase.

## Features
- Command palette launcher (⌘Space)
- On-device AI with MLX runtime
- QuickGlass contextual AI beside selected text
- Flow on-device voice dictation
- Shelf temporary file/URL working memory
- SnapWheel radial gesture menu
- Clipboard history with search (text, code, URLs, images)
- Window management and snapping
- Island system surface at macOS edge
- Notes, Kanban, Canvas, and Draw.io workspaces
- Knowledge/Recall local index search
- MCP server support
- Native Swift and JavaScript/TypeScript extension SDK
- Dock widgets
- Browser integration
- Finder actions
- Local model support (MLX, Ollama, LM Studio, Lekh AI Pro)
- BYOK cloud providers (OpenAI, Anthropic, Gemini, DeepSeek, OpenRouter)
- Menu bar cleanup (Barkeep)
- OCR clipboard intelligence
- Text snippets and expansion
- Automations and workflows
- AI personas and memory
- Color Picker, QR Code, Network Ports workspaces

## Integrations
OpenAI, Anthropic, Google Gemini, DeepSeek, OpenRouter, Ollama, LM Studio, Lekh AI Pro, MLX, MCP servers, macOS Keychain, Draw.io, Finder, Browser

## Platforms
WINDOWS, MACOS, WEB, API, CLI

## Pricing
Freemium — Free tier available with paid upgrades

## Links
- Website: https://vehla.app
- Documentation: https://vehla.app/Vehla%20Docs.dc.html
- Repository: https://github.com/ibuhs/vehla-extensions
- EveryDev.ai: https://www.everydev.ai/tools/vehla