Will GPT-5 perform better than o1 (not preview) at AIME 2024, Codeforces elo, GPQA, or the 2024 ioi?
13
Ṁ100Ṁ607resolved Aug 7
Resolved
YES1H
6H
1D
1W
1M
ALL
o1's scores are the ones mentioned here. https://openai.com/index/learning-to-reason-with-llms/
Resolves yes if OpenAI or a third party tester is able to get GPT-5 to achieve higher scores on any of these benchmarks without any kind of prompt engineering or agent scaffolding
This question is managed and resolved by Manifold.
Market context
Get 
1,000 to start trading!
1,000🏅 Top traders
| # | Trader | Total profit |
|---|---|---|
| 1 | Ṁ31 | |
| 2 | Ṁ17 | |
| 3 | Ṁ14 | |
| 4 | Ṁ14 | |
| 5 | Ṁ6 |
Sort by:
I had a previous question that asked whether I would judge GPT-5 smarter than 01 that is sitting at 85% yes.
Personally I am unsure why people think this since it is heavily implied GPT-5 won't utilize the scaled test-time compute paradigm which was so key to o1's success. I would bet that GPT-5 will be unable to do this and that the goal is more so achieving near o1 performance more quickly and economically
People are also trading
Related questions
Will GPT-5 be able to get gold on the International Mathematical Olympiad?
10% chance
Will GPT-5 be capable of achieving superhuman performance in at least one exam that is typically taken by humans?
91% chance
Will GPT-4.5 score at least 100 in an IQ test?
63% chance
Will training on o1/o3 traces improve GPT-5 performance on a broad set of tasks?
63% chance