Hints are disabled on team games.

Play other Semantle games

❤️  SEMANTLE? - Try Word Connection!


The biggest change is that we are counting DOWN for the nearest top 1000 words. Thus, the closest word to the secret word will be at 1. This seems to be more intuitive for new users. Additionally we have made some design changes. Hope you like!

Midnight UTC

Yes, .

To save your stats, login with Twitter (cloud icon above) using the browser/device you normally use, AND complete a game on this page to start.

Yes, you can now play yesterday's puzzle here.

I can think of at least four reasons for this.

  1. Your guess, or the target word, is polysemous, and the meaning that is similar is rarely used. This is why "leather" is far from "patent." Sometimes one usage is simply more popular (among newspaper reporters, which is the corpus): "display" is more often a verb than a noun, and its vector reflects this.
  2. You capitalized your word. SmartKey and some other keyboards stupidly ignore the autocapitalize settings that I have explicitly set in the HTML, and there does not seem to be anything I can do about this. I added a checkbox to help you avoid this.
  3. Your word and the target word belong to different parts of speech. Sometimes this matters a lot. Sometimes it matters only a little.
  4. By "similarity", we really mean "used in similar contexts". The principle was articulated by John Rupert Firth, who wrote, "[Y]ou shall know a word by the company it keeps." So, "love" and "hate" may seem like opposites, but they will often score similarly. The actual opposite of "love" is probably something like "Arizona Diamondbacks", or "carburetor".

The data set is what it is -- it's not perfect, and I can't afford enough computing power (or a big enough corpus) to try to make a better one. The technique has limitations. Sometimes, they'll bite you and you'll lose.

As far as old games, you can access yesterday's game. You can also play Semantle Junior, a daily, easier version of Semantle. As of May 31, you can access old puzzles at the Semantle Archives on The Word Finder. That's about it. We don't offer any other puzzles on Semantle.com. There are three reasons for this:

  1. Mindless imitation of that other word guessing game.
  2. A game designer I know found a site that let you play as much of the other game as you want. He immediately binged, and then got bored, and hasn't tweeted about it since.
  3. Another friend of mine said that she considered giving up on one day's Semantle, and then realized it would be another 24 hours before she could try again. So she persevered and got it.

But you can send your friend a custom game. And maybe your friend will send you one back. Maybe the word will be flirtatious. Maybe you will conduct an entire romance over Semantle. I met my wife through Zendo, so it's not impossible.

Prior art. I discovered French Toast about 20 years ago, and immediately knew that I had to tell Dominus about it. He replied that in fact he and Ranjit had invented it ten or so years before that, and called it "Plenty Questions". Also, the Divergent Association Task is the opposite of Semantle.

But I wasn't thinking of French Toast when I invented this. Instead, I was thinking about Wordle and Worldle and about the dimensionality of data sets. I considered Filmle, a movie guessing game: You say "The Matrix", it says "Keanu Reeves has a bacon number of 3 where bacon is the lead actor in the target film." (and so on for Carrie-Ann Moss, Laurence Fishburne, and Hugo Weaving for actors at corresponding billings). But I don't care about movies. Too bad, because IMDB actually has exactly the data that you would need to implement it in a really easy form. Then I remembered word2vec, and here we are.

GoogleNews-vectors-negative300.bin from late 2021

Unusual word found! This word is not in the list of "normal" words that we use for the top-1000 list, but it is still similar! (Is it maybe capitalized?)

No clue why the data set doesn't include them. Perhaps they were just warping the scores of everything else, since they are used in so many different contexts?

Yes, Semantle is moving to https://semantle.com. If you go to the old site, you should get redirected to the new site. This will only happen the first time you visit on a given UTC day (so we don't have to worry about transferring a day's guesses over). Hopefully, your stats and preferences should also transfer over. If you don't see your Novalis.org stats, please use this button below to migrate older stats over from novalis.org.

Semantle was created by David Turner, and is now part of The Word Finder. The Word Finder has its own word games and even solvers including an anagram solver and a Wordle Solver.

The data seems to be normalized to US spelling. Semantle tries to automatically Americanize your spelling (in the cases where only the American version is in the data set). Of course, they probably couldn't normalize some words, like 'biscuit', 'lift', or 'pants', so for those you're on your own.

Yes: click on the "#" column header. Click it again to go reverse-chronological. It will revert to similarity order when you enter a guess (or when you click the "Similarity" column header).

I was caught between a rock and a hard place here: on one hand, ranks are usually descending, but on the other hand, here, the similarities are ascending. I chose what I thought was the least bad option, but I don't think I would have been happy either way.

I don't keep statistics on this. But you are welcome to brag on Twitter, or n the subreddit. A leaderboard (a) is not in the spirit of the moment, and (b) likely to lead to unpleasant behavior.

Yes, Swedish, Hebrew, Spanish, Portuguese, French, German (or a different German one), Turkish, Russian, Dutch and Korean . Credit to enigmatix, author of Cemantix for the collaborate and authentication features.

As of May 22, you can play on a team using a code to create a room. And with a mob on Twitch, sometimes.

I grabbed a random list of the "most popular" 5,000 words in English, and removed anything capitalized or with hyphens, and the word2vec stopwords ("and", "if"). Then I shuffled it.

Yes: click the "Hint" button at the bottom of your list of guesses (to the left of "Give up"). You'll get a word that's slightly better than one of your best guesses (unless you've guessed all of the nearest thousand except the target word).