METR Q4 2026 Time Horizon - TH2 Verifiable (80%): leaderboard

METR's Q4 2026 capability update (20 November 2026): the length of task, in hours of expert human work, that a model completes 80% of the time, fitted across tasks of different lengths. Fitted only on the automatically scored half of TH2, METR's new suite: optimization contests, ML training, puzzles and games, plus the under-10-minute TH1.1 tasks METR adds to pin the short end. Much lower than the pooled series for the same model. Unbounded; higher is better. Reasoning effort and scaffold are not stated per model.

Metric: 80% Time Horizon (hours). Source: metr.org. Saturation forecast: Around December 2027. 11 models tracked. Updated 21 Nov 2026.

Top models

#ModelScoreOverall rank
1GPT-6 Astra5.44#10
2Claude Fable 5.13.16#20
3Claude Opus 5.53.04#9
4Claude Opus 53.01#26
5Claude Fable 51.99#33
6GPT-5.51.66#54
7Claude Opus 4.80.9286#65
8Claude Opus 4.70.6216#85
9GPT-5.6 Sol0.3852#37
10GPT-5.40.3346#135
11Claude Opus 4.60.1978#113

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=metr-q4-2026-time-horizon-th2-verifiable-80 · How It Works · Data refreshed daily, snapshot 2026-11-21.