Best Model on AiderBench by end of May
13
1kṀ13k
resolved Aug 9
100%98.2%
OpenAI
0.2%
xAI
0.1%
DeepSeek
0.7%
Google
0.4%
Anthropic
0.1%
Meta
0.1%
Qwen
0.1%
Other

This market resolves to the AI model that achieves the highest overall score on the AiderBench evaluation in May on the leaderboard at https://aider.chat/docs/leaderboards/. Rankings with two models will not qualify

Get
Ṁ1,000
to start trading!

🏅 Top traders

#NameTotal profit
1Ṁ993
2Ṁ136
3Ṁ129
4Ṁ128
5Ṁ71
Sort by:

@mods Manifold made this market, can this resolve?

benchmark data is stored in https://github.com/Aider-AI/aider/blob/main/aider/website/_data/polyglot_leaderboard.yml

Epoch AI has Aider Polyglot on their dashboard https://epoch.ai/benchmarks and they use release dates, unlike Aider's testing dates, which probably better matches bettor expectations, also o3 was retested after cost changes so its shown date in aider logs is June

o3 is the highest with a release date before June, slightly edging out Gemini 2.5 Pro

What happens if 2.5 Pro Deep Think isn't officially evaluated on it by the end of May, but it does in fact score higher than OpenAI?

boughtṀ100 YES

@JamesGrugett i'll buy no at 38%

© Manifold Markets, Inc.TermsPrivacy