FutureEval A
AI benchmark · Reasoning
- Publisher
- Metaculus Futureeval Official
- Tier
- A why
- Snapshot
What is FutureEval?
FutureEval measures how well AI models forecast future real-world outcomes across science, technology, health, geopolitics, AI, and other domains. Metaculus runs standardized MetacBots on most eligible Metaculus questions, normally gives them AskNews search, and compares forecasts made within the first year after each model's first forecast. The model leaderboard uses head-to-head Peer Scores based on differences in log scores and anchors GPT-4o at zero.
Tier A
Widely used: a lab reports it, or 50 or more models are scored on it.
How much stands behind a tier: usage, who backs it, a confirmed line from its own site or paper, and a clear read on X. High with three of the four, medium with two, low otherwise.
- 78 models scored
- in curated RL directories (1 pt)
On X · unclear, low buzz, from 2 posts about FutureEval in the last month
Only promotional posts from Metaculus and partner AskNews about the $50k fall tournament; no independent discussion or reported results.
How we got to tier A
- Adoption: 78 models scored. Tier A.
- On X: unclear, low buzz. “Only promotional posts from Metaculus and partner AskNews about the $50k fall tournament; no independent discussion or reported results.” No move.
- In curated RL directories: 1 point, a floor of none.
- Not yet confirmed by a line from its own site or paper.
A clear view on X moves a tier one step up or down; most entries don't move. How tiers work.
Not on the map yet: no line from its own site or paper places it in a domain.
Next: browse all benchmarks. Entry from the RL Research daily scrape of public sources, 2026-10-11 snapshot.