METR Benchmark (80% Horizon) — leaderboard
Metric: 80% Time Horizon (hours). Source: metr.org. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Mythos Preview | 3.1 |
| 2 | Gemini 3.1 Pro (Preview) | 1.5 |
| 3 | Claude Opus 4.6 | 1.16 |
| 4 | GPT-5.2 | 1.1 |
| 5 | GPT-5.3 Codex | 0.91 |
| 6 | GPT-5.4 | 0.9 |
| 7 | Gemini 3 Pro | 0.9 |
| 8 | GPT-5.1 Codex Max | 0.84 |
| 9 | Claude Opus 4.5 | 0.82 |
| 10 | GPT-5 | 0.64 |
| 11 | O3 | 0.5 |
| 12 | Claude Opus 4.1 | 0.39 |
| 13 | Claude Opus 4 | 0.34 |
| 14 | Claude 3.7 Sonnet | 0.2 |
| 15 | O1 | 0.12 |
Interactive version: theaggregate.ai/benchmark?slug=metr-benchmark-80-horizon · How the rankings work · Data refreshed daily, snapshot 2026-07-22.