METR Q4 2026 Time Horizon - Lower Lifecycle Cost Tasks (50%): leaderboard
METR's Q4 2026 capability update (20 November 2026): the length of task, in hours of expert human work, that a model completes 50% of the time, fitted across tasks of different lengths. Fitted only on the half of the pooled TH1.1 + TH2 tasks that are cheaper than their length predicts to build, run and score, by a model's cost estimate: the tasks METR expects are easier to turn into training environments. Unbounded; higher is better. Reasoning effort and scaffold are not stated per model.
Metric: 50% Time Horizon (hours). Source: metr.org. Saturation forecast: Around December 2027. 15 models tracked. Updated 21 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6 Astra | 69.64 | #10 |
| 2 | Claude Opus 5.5 | 68.62 | #9 |
| 3 | Claude Fable 5.1 | 58.86 | #20 |
| 4 | Claude Opus 5 | 31.4 | #26 |
| 5 | Claude Fable 5 | 18.63 | #33 |
| 6 | Claude Opus 4.8 | 16.1 | #65 |
| 7 | GPT-5.6 Sol | 14.09 | #37 |
| 8 | GPT-5.5 | 13.05 | #54 |
| 9 | Claude Opus 4.7 | 12.21 | #85 |
| 10 | Claude Opus 4.6 | 11.85 | #113 |
| 11 | GPT-5.4 | 11.02 | #135 |
| 12 | Claude Opus 4.5 | 7.41 | #160 |
| 13 | GPT-5 | 5.83 | #178 |
| 14 | O1 | 1.32 | #403 |
| 15 | GPT-4o | 0.1891 | #751 |
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-lower-lifecycle-cost-tasks-50 · How It Works · Data refreshed daily, snapshot 2026-11-21.