ConfBench - Confidence AUROC: leaderboard
Metric: AUROC of the verbalized confidence for separating correct from incorrect extracted entities (0-1, 0.5 is chance; verbalized top-4 guesses with probabilities in one call, OCR text with Amazon Textract confidences plus the document image, over the entity fields of 1,346 degraded variants of 75 RealKIE-FCC-Verified invoices). Source: arxiv.org. Saturation forecast: Around December 2026. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 0.84 |
| 2 | Claude Sonnet 4.5 | 0.77 |
| 3 | Claude Haiku 4.5 | 0.74 |
| 4 | Qwen 3.6 27B | 0.72 |
| 5 | Kimi K2.5 | 0.67 |
| 6 | Qwen 3 VL 235B A22B | 0.62 |
| 7 | Gemma 3 12B | 0.58 |
Interactive version: theaggregate.ai/benchmark?slug=confbench-confidence-auroc · How It Works · Data refreshed daily, snapshot 2026-09-29.