Launching today
GameOrWait
Predicts how much you'll enjoy a game before you buy it
1 follower
Predicts how much you'll enjoy a game before you buy it
1 follower
Tired of buying games you don't end up liking? GameOrWait learns your taste from your scored library and your preferences generated by the wizard, then predicts how much you'll personally enjoy any game with a target price. Free, open source, BYOK AI. Works with Steam, Playnite, GOG and others.
Hey Product Hunt 👋
I'm Amin, a dev from Berlin, and I built a tool over the past few weeks because I kept running into the same problem: I'd buy a game that reviewed well, play it for an hour or two, and realize it just wasn't for me.
Steam says "Very Positive." Metacritic says 85. Everyone raves about it, and you still bounce off it. Sound familiar?
The tool is called GameOrWait. It's free, open source, and lives at gameorwait.com.
What it does
You connect your Steam account (or import a CSV/Playnite/GOG library or manually enter games), score some games you've already played (takes a few minutes), and using the wizard in the app, set a quick taste profile, what you care about, what you hate, dealbreakers, etc.
Then when you're thinking about buying a game, you type its name and the full price, and then the AI:
Searches the web for actual reviews, current pricing, common complaints
Cross-references that against your scored library and taste profile
Returns a personal enjoyment prediction (0–100), confidence level, risk assessment, and a target price
The point is it's not giving you a generic score. It's telling you whether you will enjoy it.
Notes & Details
Free. First 5 analyses use my API key so you can try it without any setup. After that, you plug in your own AI API key (Anthropic, OpenAI, Google, or any OpenAI-compatible endpoint). Typical cost is $0.10–$0.30 per analysis. You do not pay anything to or through GameOrWait.
Privacy-first. Your API key stays in your browser's local storage. Nothing hits my servers except auth and your library and scores.
Open source under PolyForm Noncommercial. You can read the code, fork it, run it yourself.
Web + Windows desktop app (Tauri-based, native installer).
Tech stack: Next.js 16, React 19, TypeScript, styled-components, Supabase, Cloudflare Pages.
Who it's for
People who:
Have 50+ game libraries and a growing backlog
Wait for sales by default
Got burned enough times by "Very Positive" games that didn't click
Want to know if a game is worth it for them, not for the average reviewer
The honest caveats
It's a hobby project. I work on it nights and weekends.
The AI isn't magic. It's only as good as the taste profile and scored library you give it. Score at least 10–15 games across your rating spectrum or results will be noisy. The more games you score, the better results you get. The results for me personally have been pretty accurate so far.
It works best for games with decent review presence. Super obscure stuff is hit or miss, but the app also tends to that and gives you a confidence level and a score adjustment based on low confidence results.
Would genuinely love feedback, especially if the AI gives you a verdict that turns out to be wrong. Those are the most useful data points for improving the tool.