vesta-mac-dist

Overview

Vesta is a comprehensive AI chat application designed specifically for Apple Silicon Macs, featuring a robust MCP server implementation. The application supports five different AI backends simultaneously including Apple Intelligence, MLX, llama.cpp, HuggingFace Inference API, and OpenAI-compatible API servers. Its MCP server provides 41+ tools for backend management, chat operations, model handling, vision analysis, and system diagnostics, along with 6 resources and 7 prompts. The server operates on TCP loopback with token-based authentication and integrates with Claude Code to enable agentic control of the application.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• backend_management

  Manage AI backends (Apple Intelligence, MLX, llama.cpp, HuggingFace, External AI)

  Try:Switch to MLX backendTry:Configure OpenAI-compatible API server
• model_download

  Download AI models from HuggingFace or other sources

  Try:Download Qwen3-VL modelTry:Get latest Gemma 4 model
• vision_analysis

  Analyze images using vision-language models

  Try:Describe this imageTry:What objects are in this screenshot?
• conversation_search

  Search conversation history with SQL queries or full-text search

  Try:Find our discussion about macOS securityTry:Show me all messages with MLX backend
• image_generation

  Generate images using FLUX or Stable Diffusion models

  Try:Create a sunset over mountains imageTry:Generate a portrait of a cyberpunk city
• video_generation

  Generate videos from text prompts using text-to-video models

  Try:Create a 10-second video of ocean wavesTry:Generate an animation of a growing tree
• speech_to_text

  Transcribe audio to text using Whisper models

  Try:Transcribe this meeting recordingTry:Convert voice memo to text
• text_to_speech

  Convert text to speech using Kokoro or Marvis TTS models

  Try:Read this article aloudTry:Generate voice from my text
• model_management

  Load, unload, and manage AI models in memory

  Try:Unload Qwen3-VL modelTry:Check which models are currently loaded
• diagnostics

  Run diagnostic checks on the app and models

  Try:Check system resourcesTry:Test model loading performance
• ui_navigation

  Navigate and control the application UI

  Try:Switch to settings panelTry:Open conversation history view
• app_settings

  Configure application settings and backend parameters

  Try:Change MLX context sizeTry:Configure HuggingFace API key

Comparable tools

Installation

brew tap scouzi1966/afm
brew install --cask scouzi1966/afm/vesta-mac

{
  "mcpServers": {
    "vesta": {
      "command": "nc",
      "args": ["localhost", "1337"],
      "env": {"VESTA_MCP_TOKEN": "your_token_here"}
    }
  }
}

FAQ

What AI models does Vesta support?

Vesta supports 5 backends: Apple Intelligence, MLX, llama.cpp, HuggingFace Inference API, and OpenAI-compatible API servers.

How do I enable the MCP server in Vesta?

MCP server is automatically enabled when Vesta is running. Configure it in Claude Desktop with the localhost connection details.

Compare vesta-mac-dist with