vesta-mac-dist
Vesta macOS Distribution - Official releases and downloads.Vesta AI Chat Assistant for macOS - Built with SwiftUI and Apple Intelligence using Apple's on device model on MacOs Tahoe (MacOS 26). Now with side-by-side Qwen3-VL for vison
claude mcp add --transport stdio scouzi1966-vesta-mac-dist docker run -i scouzi1966/vesta-mac-dist \ --env MCP_TOKEN="optional authentication token for MCP access"
How to use
Vesta for macOS exposes an MCP server that can be controlled programmatically over a local TCP loopback connection. This MCP server lets you manage backends, initiate model loading and unloading, run conversations, query diagnostics, and navigate settings from automated clients or integration scripts. With the MCP interface you can script common workflows such as starting a Vesta session, switching active backends, and retrieving conversation history or system information, all without interacting with the GUI. If you enable token-based authentication, use an MCP_TOKEN to securely connect to the server from your client.
To use the MCP capabilities, start the Vesta Mac distribution and ensure the MCP service is reachable on the configured loopback address (typically localhost) and port. Connect from your MCP client by authenticating with the token (if enabled), then issue standard MCP commands to manage models, conversations, and settings. The server supports tools for backend management, model loading/unloading, vision analysis, and diagnostics, making it suitable for automation, testing, and advanced workflows beyond the GUI.
How to install
Prerequisites:
- macOS with Apple Silicon (recommended) or Intel hardware as supported by the app
- Optional: Docker installed if you choose the Docker-based run method
- Network access to localhost for MCP connections
Direct installation (recommended):
-
Install via Homebrew (recommended for macOS users): brew tap scouzi1966/afm brew install --cask scouzi1966/afm/vesta-mac
-
After installation, run the application to start the MCP-enabled backend. The MCP server will listen on a loopback interface for local automation scripts.
Direct download (if you prefer the DMG):
- Download the Stable or Nightly DMG from the project releases page.
- Open the DMG and drag the Vesta app to your Applications folder.
- Run the Vesta app to start the MCP server on localhost. If using a token, configure MCP_TOKEN in your client to authenticate.
Docker-based run (alternative):
- Ensure Docker is installed and running.
- Start the MCP-enabled Vesta container: docker run -it --rm scouzi1966/vesta-mac-dist
- Connect to the MCP server on localhost using the configured port and token (if enabled).
Additional notes
Tips and known considerations:
- The MCP server is documented to support token-based authentication. If you enable MCP_TOKEN, securely store the token and rotate it periodically.
- Network access is typically local (localhost). If you need remote access, configure port forwarding or a secure tunnel and update the MCP client accordingly.
- When using Docker, ensure the container has access to the necessary resources (GPU if required by backends, network permissions, etc.).
- If you encounter issues starting the MCP service, check logs from the Vesta app or container for authentication errors, port conflicts, or missing environment variables.
- The MCP interface includes tools for backend management, model load/unload, vision analysis, and diagnostics. Explore the available commands in the MCP client to maximize automation capabilities.
Related MCP Servers
Everywhere
Context-aware AI assistant for your desktop. Ready to respond intelligently, seamlessly integrating multiple LLMs and MCP tools.
chatmcp
ChatMCP is an AI chat client implementing the Model Context Protocol (MCP).
papersgpt-for-zotero
A powerful Zotero AI and MCP plugin with ChatGPT, Gemini 3.1, Claude, Grok, DeepSeek, OpenRouter, Kimi 2.5, GLM 5, SiliconFlow, GPT-oss, Gemma 3, Qwen 3.5
Risuai
Make your own story. User-friendly software for LLM roleplaying
JiwuChat
JiwuChat 🍂 : 轻量级跨平台IM聊天应用,集成AI机器人( DeepSeek/Gemini/Kimi... )、音视频通话及AI购物。支持多端消息同步,自定义主题,高效便捷 🍒
tool-ui
UI components for AI interfaces