The chess.com app is the default chess app for most people, and for good reasons that are worth stating plainly before any criticism. It has the largest player pool in the world, so a game at your level is always available in seconds. It has daily puzzles, puzzle rush, an enormous lessons library, live events, clubs, and over a hundred named bot personalities to play against. If you install exactly one chess app and never open another, that is a defensible choice and this article is not going to argue you out of it.
What it is worth being precise about is where the app stops helping, because that boundary is not where most people assume it is.
What the chess.com app genuinely does better than anything else
Three things, and none of them are close.
Finding a game. Pool size is not a feature that can be copied. At any hour, at any time control, at any rating, there is an opponent. Every other product in this space either sends you to lichess or gives you a bot.
Volume and variety of practice material. Puzzles, puzzle rush, lessons, drills, openings explorer, endgame practice. Whether all of it is well sequenced is a fair debate, but the sheer amount is real.
Bots as entertainment. Per chess.com's own Help Center, its bots and its practice-versus-computer mode are all powered by Komodo, with setting changes and custom opening books, and the app hosts more than a hundred named personalities including celebrity and creator tie-ins. Games against them are unrated. As a way to play something less punishing than a human at 11pm, it works.
Where the free account actually ends
The free tier is more generous than people expect on playing and more restrictive than they expect on understanding. Chess.com's own Help Center confirms that non-premium accounts get one full Game Review per day and that a paid tier removes that cap. Several bots show a lock icon and need an upgrade. Lessons are partly gated.
Membership tiers exist at several price points, and the higher ones add unlimited analysis and, at the top tier, coach explanations on top of Game Review. We are deliberately not printing dollar figures here: chess.com's pricing has moved, third-party sources disagree with each other by wide margins for the same tier, and the only number worth trusting is the one on chess.com's own membership page today.
The engine question, which the app answers inconsistently
This is a small thing that tells you something larger. Ask what engine the chess.com app runs and there is no single answer.
| Part of the app | Engine, per available sources |
|---|---|
| Bots and practice versus computer | Komodo, per chess.com's own Help Center |
| Game Review | Stockfish 16, server-side, per a July 2026 chess.com forum thread |
| Analysis board | Newer releases such as Stockfish 18, run client-side, per the same thread |
| Help article on computer opponents | Still says Stockfish 15, which reads as stale documentation |
None of that makes the analysis wrong. Stockfish 16 is vastly stronger than any human who will ever read this page. It does mean that when the app tells you something about your move, the provenance of that verdict is not something the app makes easy to understand, and when users ask on chess.com's own forums what analysis depth even means, the answer comes from other users rather than from the product.
If you want the underlying version of that question answered properly, we wrote a longer piece on what chess analysis can and cannot tell you, and a separate one on what Stockfish actually is.
The gap: a label is not a reason
Here is the honest shape of the problem. The chess.com app is excellent at telling you that a move was a blunder. It is much weaker at telling you why it was available to play in the first place.
Chess.com has built an answer to this: its top membership tier adds natural-language coach explanations on top of Game Review. That is a real feature and it exists today. Two things are also true about it. It sits at the most expensive tier, and chess.com's own forums carry years of active threads from paying users describing the coach commentary as vague, inconsistent, or simply absent on the moves they most wanted explained. We went through those threads in detail in our comparison of what each game review tool actually tells you.
There is a second gap, and it is bigger outside the English-speaking world: the explanation layer is English. If you play chess in Turkish, Portuguese, Spanish or French, the label translates and the reasoning does not.
The other gap: a bot is not an opponent at your level
Bot difficulty in most apps works by handicapping a strong engine, capping its skill level, search depth and thinking time. That produces an opponent that plays well and then throws in a mistake no human would make, which is why so many people describe bot games as feeling wrong without being able to say why.
Our own measurements on this are worth one number. Running the Maia3 human-prediction model against the finish of the Opera Game, with Stockfish 17 at depth 20 as the reference, we measured how likely a player at a given rating is to find the engine's own top move. In one position the engine's best move is the knight recapture 13...Nxd7, evaluated at -5.57. A 1000-rated player finds it 4.57 percent of the time. A 2200-rated player finds it 1.18 percent of the time. The stronger player is less likely to play the engine's choice, because 13...Rxd7 is the natural human recapture and better players are more consistent about playing natural moves. A throttled engine cannot reproduce that. A model trained on human games can, and we walk through the full dataset in our piece on Maia chess.
Side by side, for improvement specifically
| What you want | chess.com app | Chessdrive |
|---|---|---|
| A game right now, at your level | Best in class, largest pool in chess | Human-like bot play, plus online games; the pool is not comparable |
| Puzzles and drills at volume | Enormous library, puzzle rush, lessons | Daily puzzles tuned to your rating, smaller library |
| Post-game verdict | Accuracy score plus move labels, one full Game Review a day on the free tier | Stockfish at MultiPV=3 with move labels, plus written reasoning on every reviewed game |
| Why the move was wrong, in prose | Reserved for the top membership tier; forum threads report it as inconsistent | Included with every paid review, no higher tier to unlock |
| Explanation language | English | Your own language, across a top-20 list; notation stays in English |
| Opponent realism | Komodo with capped settings, over 100 personalities | Maia3, a model trained to predict human moves at a chosen rating |
| Events, clubs, content | Extensive | None of this |
| Platforms | Web, iOS, Android | iOS, Android |
| Cost to start | Free, with membership tiers; check chess.com's own pricing page | One free review on the house, then $5.99 a month or $39.99 a year, with a 7-day trial on annual and a 3-day trial on monthly |
So which app should you actually have
Both, probably, and that is not a dodge. Nothing here is an argument for uninstalling the chess.com app. Its playing pool, its puzzle library and its events are things a small app cannot replicate and should not pretend to.
The question is what you do in the twenty minutes after a loss. If the label plus the engine's line is enough for you, one app is enough. If what you actually want is a sentence explaining what changed three moves before the blunder, in the language you think in, that is the specific job a second app is for, and it is the one an AI chess coach is built around.
Last verified: 30 August 2026, against chess.com's own Help Center and forums as cited, checked 28 August 2026. Membership pricing, free-tier limits and feature gates on third-party products change without notice; confirm current details on chess.com's own pages before relying on them.
Frequently asked
- Is the chess.com app free?
- Yes, with limits. You can play unlimited games, do puzzles and play bots on a free account. The limits sit around analysis and lessons: chess.com's own Help Center confirms free accounts get one full Game Review per day and that a paid membership removes that cap. Some bot personalities also carry a lock icon and require an upgrade. Chess.com's pricing changes and is best read on chess.com's own membership page rather than quoted secondhand.
- What engine does the chess.com app use?
- Different parts use different engines, which surprises people. Per chess.com's own Help Center, its bots and practice-versus-computer opponents are powered by Komodo. For post-game work, a July 2026 chess.com forum thread states Game Review runs Stockfish 16 on chess.com's own servers, while the separate analysis board pulls newer releases such as Stockfish 18 run locally. A chess.com help article still says analysis uses Stockfish 15 for computer opponents, which reads as stale documentation rather than current behaviour.
- Are the chess.com bot ratings accurate?
- Treat them as orientation, not as a rating scale. Bot strength in general is set by capping an engine's skill level, search depth and thinking time, and Fairy-Stockfish maintainer Fabian Fichter, describing that same mechanism in a Lichess forum thread, called the Elo numbers attached to those levels ballpark figures for orientation rather than officially normalized ratings. A bot labelled 1400 is not a 1400-rated human, and it does not make 1400-rated human mistakes.
- Is chess.com or lichess better on mobile?
- Lichess is free with no paid tier at all, including unlimited server-side Stockfish analysis on every game. Chess.com has the larger player pool, more events, more content and a more polished app, with its deeper analysis and lessons gated behind membership. If cost matters most, lichess wins outright. If community, variety and volume matter most, chess.com does.
- Do I need a second chess app?
- Only if the thing you want is not what a big platform optimises for. Playing volume, puzzles, events and content are all covered extremely well by the chess.com app. What tends to send people to a second app is wanting a written reason a move was wrong rather than a label, wanting that in a language other than English, or wanting an opponent that plays like a human at their level rather than a throttled engine.
Keep chess.com. Add the part it leaves out.
Chessdrive reviews your games with Stockfish and an AI coach that writes out why a move was wrong, in your own language, and it plays back as a human-like opponent at your own rating. Free to start, on iOS and Android.