FutureSearch DRB - Validate Claim: leaderboard
FutureSearch Deep Research Bench task slice for validating claims against web evidence.
Metric: Average Score. Source: evals.futuresearch.ai. Saturation forecast: Around January 2027. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.6 | 0.8 | #90 |
| 2 | Claude Opus 4.6 | 0.8 | #63 |
| 3 | GPT-5 | 0.78 | #98 |
| 4 | Claude Opus 4.5 | 0.76 | #83 |
| 5 | GPT-5.5 | 0.76 | #31 |
| 6 | Claude Opus 4.8 | 0.76 | #37 |
| 7 | O3 | 0.72 | #124 |
| 8 | GPT-5.4 Mini | 0.71 | #212 |
| 9 | GPT-5.1 | 0.71 | #135 |
| 10 | Claude Opus 4.1 | 0.69 | #140 |
| 11 | Claude Sonnet 4.5 | 0.68 | #142 |
| 12 | Claude Sonnet 4 | 0.68 | #199 |
| 13 | Gemini 3.1 Pro (Preview) | 0.67 | #58 |
| 14 | Gemini 3 Pro | 0.66 | #80 |
| 15 | Grok 4 | 0.65 | #181 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=futuresearch-drb-validate-claim · How It Works · Data refreshed daily, snapshot 2026-10-20.