Will any model get above human level (92%) on the Simple Bench benchmark before September 1st, 2025.
Basic
23
Ṁ47262025
42%
chance
1D
1W
1M
ALL
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Sort by:
Is it true that this benchmark can be anything, and can be changed at any point? There are no hashes, no large sample of problems, no error bars, no evaluation code, no specifics on what a model can or cannot use... How do we know what a true performance is, except what the author says?
Description of the benchmark here: https://simple-bench.com/about.html
I have made some irrational bets to subsidize the market - as I cannot be bothered to figure out the correct way to do this.
Related questions
Related questions
80% on SWE-Bench Verified by Jan 1 2025
39% chance
Will OpenAI models achieve ≥90% on SimpleBench by the end of 2025?
38% chance
Will there be a model that has a 75% win rate against the latest iteration of GPT-4 as of January 1st, 2025?
62% chance
Will models be able to do the work of an AI researcher/engineer before 2027?
40% chance
What will be the best score on the SWE-Bench (unassisted) benchmark before 2025?
39% chance
Will simple-bench scores be reported in a major AI lab paper or blog post by the end of 2024?
52% chance
Will an AI model outperform 95% of Manifold users on accuracy before 2026?
56% chance
Will any foundation model score 75 or higher on The Foundation Model Transparency Index by the end of 2024? (LLama2: 54)
45% chance
Will an AI SWE model score higher than 50% on SWE-bench in 2024?
20% chance
What year will the first AI exceed 80% on MLE-bench?