LiveBench Long Horizon: leaderboard
Metric: Score. Source: livebench.ai. Saturation forecast: Around October 2027. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 85 | #3 |
| 2 | GPT-6 Astra | 82.4 | #5 |
| 3 | Claude Fable 5.1 | 80.8 | #8 |
| 4 | Claude Sonnet 5.5 | 78.8 | #11 |
| 5 | GPT-6.1 Sol | 73 | #9 |
| 6 | Gemini 4 Argon | 69.9 | #7 |
| 7 | GPT-6 Sol | 66.9 | #16 |
| 8 | Claude Opus 5 | 62.1 | #12 |
| 9 | Claude Haiku 5.5 | 56.6 | #28 |
| 10 | Claude Fable 5 | 53.2 | #15 |
| 11 | Grok 4.7 | 50.1 | #26 |
| 12 | GPT-5.6 Sol | 48.5 | #18 |
| 13 | GPT-5.6 Terra | 48.4 | #42 |
| 14 | Claude Opus 4.8 | 48.3 | #36 |
| 15 | GPT-6 Luna | 48.1 | #51 |
No result here: #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=livebench-long-horizon · How It Works · Data refreshed daily, snapshot 2026-10-14.