BrowseComp (Mercor 130): leaderboard
Metric: Pass@1 (%). Source: www.mercor.com. Saturation forecast: Around January 2027. 40 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-6.1 Sol (Max) | 95.4 |
| 2 | GPT-6 Astra (xHigh) | 94.2 |
| 3 | GPT-5.6 Sol (xHigh) | 90.6 |
| 4 | GPT-6 Sol (Max) | 89.4 |
| 5 | Kimi K3 (Max) | 89 |
| 6 | Claude Opus 5.5 (Max) | 88.5 |
| 7 | Claude Sonnet 5.5 (Max) | 86.6 |
| 8 | GPT-5.6 Terra (Max) | 85.8 |
| 9 | DeepSeek V4.1 Flash (Max) | 85.8 |
| 10 | Claude Fable 5.1 (Max) | 85.2 |
| 11 | Claude Opus 5 (Max) | 84.6 |
| 12 | Grok 4.6 (xHigh) | 84 |
| 13 | GPT-5.6 Luna (Max) | 83.8 |
| 14 | GPT-6 Luna (Max) | 83.3 |
| 15 | Gemini 3.8 Flash (High) | 83.1 |
Interactive version: theaggregate.ai/benchmark?slug=browsecomp-mercor-130 · How It Works · Data refreshed daily, snapshot 2026-10-09.