Choose theme
AI Visuals
Genex Desktop
Genex is a desktop workspace for making browser games with AI. Describe a game or a change in chat, then explore the playable result, build history, and assets beside the conversation.
The game remains a local project you can open, edit, and export. The project describes Genex as early, with rough edges to expect. Use this as a first read, not a recommendation. Open the original project before trusting details like terms, limits, privacy, cost, setup, or safety.
What it is
An AI workspace for browser games
The desktop app brings a game conversation together with Live, Builds, and Assets views. It supplies the workspace around your chosen model, including game files, a playable preview, and export.
Why it stands out
Separate models for different jobs
Loop mode can delegate work while the main conversation leads. Main agents, workers, and reviewers have separate model selections, so one game-building workflow can use different providers for those roles.
Availability
Early desktop releases and public source
GitHub releases provide downloads for Apple Silicon macOS and x64 Linux. Windows support is listed as coming soon. The repository declares MIT; use the repository for current terms and download choices.
Why it matters
What makes it useful
For a small game idea, you can describe the mechanic, try the resulting game in Live, and request a change from the same conversation. For example, a follow-up about a jump or a control can stay with the project you just played. The source describes a workflow for reviewing running games and build results; an AI-generated result still needs your judgment about how it plays.
What to know
Where it fits
Genex produces local browser games. Export creates a static web bundle; hosted publishing through the Genex plugin is a separate route with account and service requirements. Games can be built, previewed, and exported without plugins. Unity and Unreal plugins, along with native C++ games, are listed as planned rather than current capabilities.
Notable points
What stands out
An asset appearing in the library is a different milestone from that asset working in the game. Assets can preview images, audio, video, and animated 3D models, while Builds records work and reviewer notes. The documentation explicitly separates a finished worker, passing checks, integration, and the Live revision. That distinction helps you read a build result: a generated model or completed task alone does not establish that the delivered game uses it correctly.
Before using
What to review
Claude Code and Codex need their own installed CLI and sign-in. Local routes include Ollama and the documented Apple Silicon Bonsai option; model and subscription limits still apply.
The bundled Local Blender plugin needs Blender. Genex asset generation and hosted publishing use a Genex account and paid credits, separate from basic game creation and export.
The privacy document says the selected model receives conversation context, tool results, and relevant files or screenshots. A local game folder does not mean a cloud-model workflow stays on your computer.
The privacy document describes native plugins as executable code with your access. The plugin guide describes their child process as crash isolation, rather than an operating-system sandbox, and says the install dialog names the publisher, capabilities and scan verdict.
Reader fit
Who may find it relevant
People exploring small browser-game ideas can inspect a workflow that keeps the conversation near the game they can play. Builders can open the generated project in their own editor or extend the app with tools and plugins. Expect model setup and early-app friction; anyone seeking a current Unity or Unreal integration should keep that requirement separate from the browser-game workflow available here.
Editorial note
Why LifeHubber lists it
We list Genex because its playable preview and editable local project give a game idea somewhere to go beyond the chat reply. For someone learning by changing a mechanic and trying it again, the conversation stays beside the thing being made, while export provides a way to take the result elsewhere. That connection between asking, playing, and keeping the project is the reason to explore it; the model you choose and the quality of its game still matter.
Source links
Source materials
Reader note
Before relying on this entry
LifeHubber lists entries to help readers inspect AI projects, not to endorse them or prove they are safe, suitable, accurate, maintained, or right for a specific use. We do not verify every entry in depth. Before relying on anything listed, review the original materials, terms, privacy practices, limits, and risks that matter for your situation.
More in AI Agents
Keep browsing this category
Explore more AI agent projects.
Paperclip
paperclipai/paperclip
A self-hosted server and dashboard for coordinating agent teams through companies, goals, roles, issues, heartbeats, budgets, approvals, and persistent activity records.
Octop
TencentCloud/Octop
A self-hosted assistant platform with multiple users and experts, shared knowledge bases, messaging channels, scheduled tasks, workspace memory, and configurable model providers.
Last30Days Skill
mvanhorn/last30days-skill
A cross-client agent skill for 30-day topic briefs across community, market, code, video, social, web, and optional local-file sources, with several install paths, source diagnostics, source-attribution rules, and a Python 3.12+ runtime.
For project maintainers
Listed here? You can use the badge.
If you maintain a project with a current LifeHubber listing, you may add the optional “Listed on LifeHubber AI Resources” badge to its README, docs, or website. No introduction or permission request is needed.
