ModelSweeper

Interface Preview

ModelSweeper is a desktop command center for AI model workflows. It helps you choose the right models, validate memory fit before runtime, and keep role-based presets synchronized across tools.

Why ModelSweeper

• Reduce trial-and-error before inference starts
• Switch role presets quickly (Thinker, Coder, Reviewer)
• Estimate RAM/VRAM fit with practical hardware inputs
• Keep local and cloud-oriented workflows in one clean UI

Live Links

• Website: blackwolf9-coder.github.io/modelsweeper
• Releases: github.com/Blackwolf9-coder/modelsweeper/releases
• Launch Plan: LAUNCH_PLAN.md
• Campaign Kit (EN): CAMPAIGN_KIT.md
• Campaign Kit (AR): CAMPAIGN_KIT_AR.md
• Product Hunt Pack: PRODUCT_HUNT_READY.md
• v0.2.0 Roadmap: ROADMAP_v0.2.0.md

Core Features

• Role-based presets with instant activation
• Model catalog with one-click role assignment
• Live system RAM tracking vs selected preset load
• Hardware Calculator with slider-based controls:
  • Model size (B params)
  • Context length (tokens)
  • System RAM (GB)
  • VRAM per GPU (GB)
• Integration settings export for external clients

Product Areas

• Dashboard: machine RAM state and preset load warnings
• Models: manage installed models and assign operational roles
• Preset Builder: compose simultaneous/sequential execution profiles
• Hardware Calculator: estimate RAM, VRAM, and deployment fit
• Integrations: map active preset settings to tool integrations

Stack

• Electron
• React + TypeScript
• Zustand
• Vite
• Tailwind CSS

Getting Started

Requirements

• Node.js 18+
• npm 9+

Install

npm install

Run (Dev)

npm run dev

Validate

npm run typecheck
npm run lint

Build

npm run build

Package Desktop App

npm run dist

Contributing

See CONTRIBUTING.md.

Security

See SECURITY.md.

License

MIT