Cubric Vision Overview

Start here if you want the shape of the app before you dive into install steps, models, projects, prompts, queueing, history, and tools.

Pick a route from the sidebar when you need a specific answer, or use the video below as a guided tour before reading the full docs.

What is Cubric Vision?

Cubric Vision is a desktop app for local AI image and video work. It wraps ComfyUI as the generation engine, but the working surface is organized around projects, media, prompt controls, queueing, and history instead of node graphs.

The app keeps projects on your machine. Model files live separately in the local engine folders, while project folders carry the media, history, settings, and metadata for the creative work itself.

New in version 1.4

The largest release since 1.0. The headlines:

• The canvas toolkit — the mask tools rebuilt as a group with ten brushes, shapes, and an Adjust that grows, shrinks, or outlines a mask; a new Paint group; and Composite, which blends two of your images while you watch. See Image Tools.
• Control — one operation that copies the structure of a reference image, with a Control Type picker and a strength slider. It replaces Depth everywhere, and Chroma gains it.
• Localised edit — mask an area before an edit and only that area is regenerated, at the picture's own resolution.
• MiniMax H3 and MiniMax H3 Reference — video that comes with its own synchronized sound, and a sibling that keeps the same character, place, or voice across clips with no training. See Models.
• Animate towards an image — a dropped still can now be the clip's last frame, not just its first.
• A rewritten engine setup — engine updates apply in place in seconds instead of an eleven-gigabyte re-download. See Getting Started.

New in version 1.1

Version 1.1 is a large update. The headlines:

• RunPod remote engine — run generation on a rented cloud GPU while your work stays local until you connect. See Settings.
• Full-page Model Library and a wave of new models — Krea 2, Qwen Image Edit, Boogu Image Edit, Chroma, LTX 2.3 (with sound), NVIDIA PiD Upscaler, and Wan 2.2 5B. See Models.
• New History tools — Remove Background and Export GIF.
• Redesigned Settings, bring-your-own LoRA, per-operation installs, and a broad quality-of-life sweep across the app.

Watch the overview

This video is nearly an hour long. Use it as an alternate starting point if you prefer to watch the product flow first. The written docs remain more complete for specific controls, edge cases, screenshots, and reference details.

Two shorter walkthroughs cover the recent releases and remote setup:

How the app flows

The main path is Landing to Gallery to History. The Landing page is where you create or open a project. The Gallery is the default project workspace, where media cards, prompt controls, model choices, and the Cue queue come together. History opens from a Gallery card when you want to inspect one image or video group and continue from a specific result.

The Prompt Box appears in the Gallery and History workspaces. It snapshots the selected model, operation, prompt text, media inputs, and controls when you press Cue. Loop builds on the same Cue path when you want the app to keep generating from the current Prompt Box state.

What to read next

• Start with Getting Started if you need the portable release, first-launch setup, model folder, and Settings path.
• Use Projects when you need to understand where project data lives and how project folders move between machines.
• Use Prompt Box and Queue when you are learning the generation loop, Cue, Loop, model settings, media inputs, and resumable video controls.
• Use History, Image Tools, and Video Tools when you are refining an existing result.
• Use Models and Hotkeys for install states, operation support, and keyboard shortcuts.