ConfBench - Error Capture (ECARB@30): leaderboard
Metric: Error capture at a 30 percent review budget (ECARB; errors found by reviewing the least-confident 30 percent of predictions divided by the errors a random 30 percent review finds, so 1 is no lift and 3.33 the ceiling; 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 | 2.43 |
| 2 | Claude Sonnet 4.5 | 2.24 |
| 3 | Claude Haiku 4.5 | 1.96 |
| 4 | Qwen 3.6 27B | 1.93 |
| 5 | Kimi K2.5 | 1.68 |
| 6 | Qwen 3 VL 235B A22B | 1.53 |
| 7 | Gemma 3 12B | 1.39 |
Interactive version: theaggregate.ai/benchmark?slug=confbench-error-capture-ecarb-30 · How It Works · Data refreshed daily, snapshot 2026-09-29.