vibedgame is a daily launch board for browser games. Makers post something they finished, people play it and vote, and the best of the day rises. No installers, no launcher, no forty-gigabyte download to find out you do not like it.
Every day a handful of games launch here. You open the front page, you see what went up, and every card is one click away from a game that is already running in your browser. There is no store page in between, no wishlist, no “add to library”. If a game looks interesting you are playing it four seconds later, and if it turns out to be bad you close the tab and you have lost nothing.
That is the entire product. Everything else on the site (categories, collections, the leaderboard, the launch archive) exists to help you find your way back to a game you liked, or to find the next one.
Why it exists
Browser games have a distribution problem that has nothing to do with quality. They do not fit anywhere. A Steam page assumes a download. An app store assumes an install. A jam page assumes you already knew about the jam. The result is that an excellent thirty-minute game made by one person in a fortnight has roughly the same chance of being seen as a tweet at 3am, and a much worse chance than a screenshot of it.
Meanwhile the appetite is obvious. People play games at their desk on purpose. They want something that starts instantly, ends cleanly, and does not ask for an account or a credit card or a Tuesday commitment. That audience is large and badly served, and the games it wants already exist. They are just scattered across a hundred personal domains and a thousand jam entries with nine views each.
Share
So: one page a day, ranked by people who played the games. It is not a complicated idea. It works because the constraint is real. Everything here has to load in a tab, which is a much better filter than any curation policy we could write.
How ranking works
The front page is sorted by a hot score, which is a number every published game carries and which decays with age. Here is the whole thing, exactly as it runs:
signal = votes + (comments × 0.34)
hot = (signal + 1) / (hoursSinceLaunch + 2) ^ 1.5
Three decisions are buried in that, and they are all arguable, so they are worth stating plainly.
Comments count for about a third of a vote. A game that starts a real conversation is doing something a quiet game with the same vote count is not. Weighting comments at 1.0 would turn the front page into an argument board; weighting them at zero would bury the strange, divisive games that are often the most interesting thing on the site. 0.34 is a judgement call, not a discovery.
Age is measured in hours and punished at a 1.5 exponent. Two hours after launch, the denominator is 8. A day later it is about 130. That means a game needs roughly sixteen times the votes to hold the same position a day later, which is the point. The front page is supposed to turn over. If it did not, the same five games would sit there for a month and nothing new would ever get played.
The +1 and the +2 stop new games from being ranked by luck. Without them, a game with one vote in its first minute would outrank everything on the site. The constants flatten the first hour so that a game has to actually accumulate something before it climbs.
Things the ranking deliberately does not do:
There is no paid placement. Not now, and if that ever changes it will be labelled on the card itself, not buried in a disclosure page.
Nobody can hand-boost a game into the hot list. Editorial taste lives in collections, which are clearly marked as somebody’s opinion and sit in their own part of the site.
Votes are not weighted by who cast them. An account with 4,000 karma and an account made this morning move a game by exactly the same amount.
There is no personalisation. The front page you see is the front page everybody else sees. That is the point of having a front page.
What gets removed
Very little, and always for a reason we can name. The full policy is in the submission rules, but the short version: a listing comes down if the link is dead, if the game is not yours, if it is not actually a game, if it is gambling with real money, or if the votes on it were bought or coordinated.
Vote manipulation is the one we take seriously, because it is the only thing on this list that damages everyone else’s launch. A ring of accounts voting for each other does not just inflate one game. It pushes four honest games off the front page. When we find it, the votes are reversed, the game goes back into the queue, and the accounts are gone. There is an appeal process and we do read the appeals.
A game being bad is not a removal reason. Neither is a game being small, unfinished-looking, weird, ugly, or five minutes long. Those are ranking problems, and the ranking already handles them.
The AI question
A lot of the games here were built with help from AI tools. Makers credit what they used at submission, and those credits become the Built with filter, so you can see every game made with Claude Code, or Phaser, or nothing but the canvas API.
That is all it is: a factual note about process, the same as listing an engine. It is not a badge, it does not affect ranking, and a game does not get graded on a curve for it. The credits are self-reported and we do not audit them, because the honest answer for most projects in 2026 is “some of it”, and pretending there is a clean line would be worse than the mild imprecision of letting people describe their own work.
Who it is for
If you play games: you have ten to thirty minutes and you would rather spend them playing something than reading about it. You do not want a launcher. You are fine with a game being short. You are the reason the play button is the biggest thing on every card.
If you make games: you finished something small and it is sitting on a URL nobody has visited. You do not have a marketing budget, a publisher, or a mailing list, and you would like the first hundred people to find it without you having to beg in six Discords. Launching is free, takes about ten minutes, and you keep every right to your game. See the terms for the exact licence, which is narrow on purpose.
If you are neither: the daily email is five games and nothing else. It is the lowest-effort way to use this site.
Who runs it
A small team, publicly. Site changes are written up in the changelog as they ship, including the ones that did not work. If something here is broken, unfair, or just annoying, hey@vibedgame.com reaches a person, and moderation questions get answered by the person who made the call rather than a form.
The site has been running since 2026. It is not venture-funded, it does not run ad networks, and it does not sell your reading history. The privacy page lists everything we store, which is a short list.