Will a model achieve a METR 50% time-horizon of 4+ hours by the end of 2025?
27
Ṁ2110Dec 31
29%
chance
1D
1W
1M
ALL
METR's 50%-time horizon doubling time trendline from more recent models indicates that a model released towards the end of 2025 will likely have a time horizon of over 4 hours. Will this come true?
See links for more details:
https://www.alignmentforum.org/posts/SuvWoLaGiNjPDcA7d/metr-s-evaluation-of-gpt-5
https://www.lesswrong.com/posts/deesrjitvXM4xYGZd/metr-measuring-ai-ability-to-complete-long-tasks
This question is managed and resolved by Manifold.
Get
1,000and
3.00
1,0003.00
Related questions
Related questions
Will a single model achieve superhuman performance on all Atari environments by 2025?
5% chance
Grok 4.20's METR 50% time horizon
Will a Google model lead METR's task duration chart by EOY?
26% chance
Best AI time horizon by August 2026, per METR?
Best AI time horizon by February 2026, per METR?
Will GPT-5.2's METR 50% time horizon exceed 3 hours 30 minutes?
29% chance
Will the METR long-horizons have a >6 month doubling time for at least a 4 month period before 2026?
60% chance
R2's METR 50% time horizon
Will GPT-5.2 surpass 3 hours in METR time horizon?
74% chance
Will I be able to complete a 10+ hour trip via self-driving car by end of 2025?
3% chance