IMA2 Gen Free and Open Source Desktop App CookApps CookApps Marketplace

Overview

IMA2 Gen (ima2-gen) is a comprehensive, local-first visual generation studio and runtime engineered for creators and autonomous coding agents. Storing every asset privately in ~/.ima2/generated, it unites ChatGPT (OAuth/API), xAI Grok (Image & Video), **Google Gemini (nano-banana-2), NovelAI, MiniMax, AtlasCloud, and local ComfyUI** pipelines inside a polished browser UI, Apple-notarized macOS desktop app, Windows installer, and composable CLI.

How It Works

IMA2 Gen operates through structured, deterministic execution stages:

Multi-Provider Visual Ingestion: Routes prompts and reference images across ChatGPT OAuth/API, xAI Grok (text-to-video), Google Gemini, MiniMax, and local ComfyUI workflows.

Branching Node Graph: Preserves parent image ancestry in an interactive node graph, enabling users and coding agents to push visual concepts in multiple parallel directions.

Canvas Mode with Alpha Verification: Provides an infinite canvas with pan, zoom, eraser, sticky notes, background cutout, and automated server-side PNG alpha transparency verification.

Vector Tracing & Agent Skills: Converts raster compositions into clean SVG vector paths via @neplex/vectorizer while serving structured Markdown skills for autonomous AI coding agents.

IMA2 Gen prompt studio workspace with reference images and model lane settings

IMA2 Gen node graph branching source image into parallel artistic variants

IMA2 Gen canvas mode with eraser, background removal, and verified alpha transparency

IMA2 Gen home workspace displaying provider lane readiness and recent generations

Key Technical Features & Capabilities

Multi-Provider Image & Video Lanes: Generates and edits images across GPT-6/gpt-image-2, Gemini, NovelAI dual prompts, MiniMax, and Grok text-to-video / image-to-video with live SSE progress tracking.

Branching Node Graph & Reference Composer: Attaches up to 5 reference images (14 for video) and branches successful visuals in parallel parent-child trees without overwriting prior results.

Canvas Mode & Verified Alpha Cutouts: Features zoomable annotation canvas, background cutout, sticky notes, alphaVerified server-side transparency checking, and @neplex/vectorizer SVG tracing.

Agent-Native Markdown Skills & CLI: Exposes ima2 gen, ima2 video, ima2 vectorize, and installable Core/Frontend/UIUX design skill modules (ima2 skill install) for AI coding assistants.

Architecture & Technology Stack

Primary Languages: TypeScript, Node.js 22+

Frameworks & UI Runtime: Local Web Studio (:3333), Electron macOS & Windows Desktop Wrapper, React UI

Native Media & Data Engine: better-sqlite3, sharp, @neplex/vectorizer (Raster-to-SVG)

Supported Platforms: macOS Apple Silicon (Signed & Notarized DMG), Windows (x64 & ARM64 EXE), Linux (.AppImage, .deb), Docker

System Requirements

Specification	Minimum Requirement	Recommended Specification
Operating System	macOS (Apple Silicon DMG & npm), Windows 10/11 (x64 & ARM64 EXE), Linux, Docker	Latest stable release build
Processor Architecture	4-Core x86_64 or Apple Silicon M1	8-Core Apple Silicon / Core i7+
Memory (RAM)	4 GB Unified Memory	16 GB+ High-Speed RAM
Disk Space	1 GB free SSD storage	5 GB+ free SSD storage

Installation & Setup Guide

5 Redit

Desktop App & CLI Installation

macOS Desktop: Download ima2-3.26.3-mac-arm64.dmg (Apple Developer ID notarized, bundled runtime — no Node.js needed).

Windows Desktop: Download ima2-3.26.3-win-x64.exe (or win-arm64.exe).

CLI / Local Web Studio (Node 22+):

bash

npm install -g ima2-gen
ima2 setup
ima2 serve   # Opens http://localhost:3333

One-Line Installers

macOS: curl -fsSL https://ima2gen.com/install-mac.sh | bash

Windows (PowerShell): irm https://ima2gen.com/install-windows.ps1 | iex

Installation & Setup Guide

Mở khóa ứng dụng để xem chi tiết hướng dẫn cài đặt, cấu hình môi trường và vận hành.

How to Use & Workflow

5 Redit

Authenticate Provider Lanes: Run ima2 setup or connect ChatGPT OAuth, Grok OAuth, Gemini API keys, or local ComfyUI endpoints in Settings.

Generate & Branch Visually: Compose prompts with reference images in Create, or split results into parallel directional branches inside Node Graph.

Edit & Vectorize on Canvas: Remove backgrounds with server-verified PNG alpha checks, trace crisp SVGs via ima2 vectorize, or compose video storyboards.

Automate via CLI & Agent Skills: Pin model defaults (ima2 defaults set image oauth/gpt-6-luna) and install coding agent skills via ima2 skill install.

How to Use & Workflow

Mở khóa ứng dụng để xem chi tiết hướng dẫn cài đặt, cấu hình môi trường và vận hành.

Troubleshooting & Common Issues

**Issue: CLI generation fails with NO_DEFAULT_MODEL.**

*Cause*: Fail-closed safety policy refuses silent provider fallbacks until an explicit default model lane is saved.

*Resolution*: Run ima2 models and assign defaults using ima2 defaults set image <lane>/<model> or pass --model <lane>/<model>.

Issue: Docker container refuses non-loopback requests or hangs on setup.

*Cause*: Server enforces IMA2_LAN_TOKEN security header when bound to 0.0.0.0.

*Resolution*: Pass -e IMA2_LAN_TOKEN=<secret> in Docker and run node server.js instead of interactive ima2 setup.

IMA2 Gen

Overview

How It Works

Key Technical Features & Capabilities

Architecture & Technology Stack

System Requirements

Installation & Setup Guide

Desktop App & CLI Installation

One-Line Installers

Installation & Setup Guide

How to Use & Workflow

How to Use & Workflow

Troubleshooting & Common Issues

How to Use & Getting Started

How to Use & Getting Started

Ready to run IMA2 Gen?

More in graphic-design

VectorCraft

LightCraft

PhotoCraft