Criteria suggestions for a deterministic chess bot
Ṁ80 / 150
bounty left

For some questions I am drafting, and also for some personal programming projects/my blog, I have been looking for a chess-playing agent agent with the following properties:

  • It evaluates any position in about 1-10 seconds worth of time, running on a laptop.

  • It gives roughly accurate evaluations, about as good as can be expected for a chess engine running with this amount of time, using the usual scoring system.

  • It is easily downloadable/runnable by me/anyone else with a computer.

  • It is deterministic: It gives the same evaluation/top move in any particular position, no matter what hardware it is running on or what positions it has evaluated recently.

Stockfish meets the first three, but I am not sure about how I can get it to be deterministic (running it single-threaded?). I would definitely be interested in command-line arguments for stockfish that guarantee a deterministic result if they exist. beyond that, any other suggestions would also be appreciated.

Chess
Get
Ṁ1,000
to start trading!

People are also trading

Which of these language models will I beat at chess?
Which of these Language Models will beat me at chess?
Will an AI image generation model successfully generate a proper chess board by July 31, 2025
12% chance
Before 2026, a publicly available AI system highly reliably screens videos of chess games for touch move violations?
40% chance
Will GPT-5 have a rating of at least 2000 in chess?
38% chance
Would an AI be capable of constructing a non-brute, ultraweak proof of chess for this board state by 2030?
49% chance
What values of (states * steps) suffice for a Turing machine that weakly solves chess?
Will we get "chess AGI" by 2030?
80% chance
In 2028, will I use a chatbot that can win >25% of Turing test games (defined within) where I am the judge?
35% chance
Will a Language Model under 10B parameters play chess at Grandmaster level by 2050?
88% chance
© Manifold Markets, Inc.•Terms•Privacy