Stockfish
No written verdict yet. Here is what commenters said.
Why people recommend it
“Stockfish 19 has gotten better at understanding fortresses”
“To show you how strong Stockfish 19 is compared to 18, I used to lose 100% against 18, and I now lose 100% against 19”
Show 3 more
“Now I can use an even better Stockfish to benchmark my own engine”
“Go throw yourself at Stockfish or another chess engine for this. They tend to have good test harness”
“using Stockfish with multipv”
Caveats and dissent
Show 2 more
“the set of valid moves you are passing into Stockfish only affects the initial move.”
“where Stockfish blunders a seemingly winning position”
Often compared with
From the makers not counted in the ranking
“it's not only my change is making stockfish not blunder”
Threads where it came up
- Stockfish 19 10 mentions
- Chess engines do weird stuff 4 mentions
- Small Models Have Arrived 3 mentions
- Show HN: Constrained Chess, Play Stockfish with custom natural-language rules 3 mentions
- Show HN: A Claude Code skill to analyze your chess games 2 mentions
- Regex Chess: A 2-ply minimax chess engine in 84,688 regular expressions 2 mentions
Data: stockfish.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 40
Why don't you mention the second match here, with its adjustments to meet Stockfish's quibbles - and the same result? https://en.chessbase.com/post/the-full-alphazero-paper-is-published-at-long-last/1
“integrates an LLM of your choice with StockFish to analyze games”
“I tried to use claude to analyze my games using stockfish”
“they shit out a carbon copy of https://github.com/official-stockfish/stockfish”
Shameless self promotion: If you’re a Mac user and want something with native look and feel, there is also The Marvelous Brass Chessplaying Automaton (TestFlight beta): https://testflight.apple.com/join/mqfhpmHY There are also iOS and watchOS versions, but those can’t run 3rd party engines like Stockfish.
“Stockfish 19 has gotten better at understanding fortresses”
“Stockfish was first, Reckless v0.9.0 was second”
> I also see the issue on Lichess’s GUI by going to this URL: > https://lichess.org/lwiPq9wB#48 When I go to that link it shows Black is winning with -1.9 eval at depth 32 with SF19 1MB NNUE! It looks like it got this evaluation from the cloud database.
“stockfish will sometimes take a bad read from a cache”
“Interesting bug or regression.”
I can't beat the famous 1k JavaScript version either. I was better when I was younger. https://nanochess.org/archive/toledo_javascript_chess_3.html
“To show you how strong Stockfish 19 is compared to 18, I used to lose 100% against 18, and I now lose 100% against 19”
“Now I can use an even better Stockfish to benchmark my own engine”
> a large language model trained on chess commentary as well as being trained on millions of chess games would outperform stockfish which is only trained on millions of chess games Not really, if anything it's closer to the opposite. The Bitter Lesson essay literally has this as an example: > These researchers wanted methods based on human input to win and were disappointed when they did not.[1] and > Enormous initia
“in 2020 Stockfish 12 adding some NN evaluation”
“https://mcpmarket.com/server/stockfish”
“against "diluted" versions of Stockfish”
“In the Stockfish notation this engine uses”
“Stockfish's heuristic for "risk of zugzwang" is basically "only kings and pawns left over"”
“Go throw yourself at Stockfish or another chess engine for this. They tend to have good test harness”
“integrate Stockfish like Engine for Baghchal one day”
“https://www.chess.com/news/view/stockfish-outlasts-nakamura-3634”
“Nakamura played with Rybka against Stockfish from 2014”
> I knew that it was just a minmax with alpha beta pruning and a really good eval function. Now its not. It is still "just" a minimax with alpha beta pruning, except the eval function is now a neural network. NNUE, to be more specific. I highly advise anyone who is curious about chess engines, but hasn't heard about NNUE to read about it. I find this technology absolutely fascinating. The key idea is that a neural ne
“Here's an example of me beating Stockfish 18 at depth 1”
“It uses stockfish and AI for analysis.”
“Stockfish can never actually play against it”
“the set of valid moves you are passing into Stockfish only affects the initial move.”
“It'd be nice if stockfish or its variants exposed the "ply searcher"”
“stockfish's WDL is very rudimentary”
“Stockfish wins the rematch also.”
“Here's a game from a month ago where Stockfish loses to Lc0”
“Stockfish does lose every now and then against top engines.”
“Stockfish is not unbeatable by another engine, or another copy of Stockfish.”
“Stockfish 18 is estimated to be in excess of 4000 elo”
“Play against Stockfish”
“using Stockfish with multipv”
“I'm running games through stockfish/lc0/Maia”
“These Stockfish multipv (multi-principal-variation) outputs show something interesting”
“where Stockfish blunders a seemingly winning position”