Vals AI ClaimBench: leaderboard
ClaimBench v1: 1,960 claims paired with 10-K excerpts; three-way supported, contradicted or insufficient-evidence label accuracy. Five construction categories have roughly 392 claims each. This is the full label-accuracy table, not selective coverage or the threshold-tuned cascade on a 70% holdout. Public examples are available.
Metric: Accuracy (%). Source: www.vals.ai. Saturation forecast: Estimated already saturated. 19 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Sol (Max) | 100 | #8 (GPT-6.1 Sol) |
| 2 | Claude Haiku 5.5 (Max) | 100 | #40 (Claude Haiku 5.5) |
| 3 | Claude Opus 5.5 (Max) | 100 | #4 (Claude Opus 5.5) |
| 4 | Claude Sonnet 5.5 (Max) | 100 | #11 (Claude Sonnet 5.5) |
| 5 | GPT-6 Luna (Max) | 99.18 | #59 (GPT-6 Luna) |
| 6 | Jev 1.13 | 97.65 | #316 |
| 7 | GPT-6 Luna | 96.89 | #59 |
No result here: #6 GPT-6 Astra, #7 Gemini 4 Argon, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=vals-ai-claimbench · How It Works · Data refreshed daily, snapshot 2026-10-21.