Which AI model will win the Kaggle Game Arena Chess tournament?
47
Ṁ1.1kṀ23kresolved Aug 8
100%99.0%
o3
0.1%
o4-mini
0.1%
DeepSeek-R1
0.2%
Gemini 2.5 Pro
0.1%
Claude Opus 4
0.3%
Grok 4
0.0%
Gemini 2.5 Flash
0.0%
Kimi K2
0.1%Other
This question is managed and resolved by Manifold.
Market context
Get
1,000 to start trading!
🏅 Top traders
| # | Trader | Total profit |
|---|---|---|
| 1 | Ṁ832 | |
| 2 | Ṁ446 | |
| 3 | Ṁ297 | |
| 4 | Ṁ146 | |
| 5 | Ṁ129 |
People are also trading
Which company has the best AI model end of February 2026? (LMArena)
Will a large language model beat a super grandmaster playing chess by EOY 2028?
43% chance
Which of these Language Models will beat me at chess?
Will an AI by OpenAI beat a super grandmaster playing chess by 2028?
57% chance
Which chess engine will be the strongest at the end of 2028?
Will OpenAI model win first "inter-AI-model diplomacy" game where the game is EU4/5, Civ6/7, or AoE2 Regicide Rumble?
50% chance
Sort by:
they should’ve done it in PGN format since that greatly increases the legal move rate (was like 99.9% with 3.5 turbo instruct) then have separate channels for commentary
instead of allowing 4 attempts and feedback if a move is illegal.
Not sure about the exact scaffold they’re using but it’s clearly leading to low quality chess when a model from over 2 years ago was 1800.
@ChinmayTheMathGuy I don't believe any model was ever 1800, but even those that do believe it - nothing has come close since
@FergusArgyll there was a site called parrot chess and it was pretty good. I would say around 11600 just from playing against it
People are also trading
Related questions
Which company has the best AI model end of February 2026? (LMArena)
Will a large language model beat a super grandmaster playing chess by EOY 2028?
43% chance
Which of these Language Models will beat me at chess?
Will an AI by OpenAI beat a super grandmaster playing chess by 2028?
57% chance
Which chess engine will be the strongest at the end of 2028?
Will OpenAI model win first "inter-AI-model diplomacy" game where the game is EU4/5, Civ6/7, or AoE2 Regicide Rumble?
50% chance