Doc2DB-Bench - Multi-Entity Binding: leaderboard

Metric: Cell-level F1 (%; relation-table cells annotated with the Multi-Entity Binding capability of the inter-table taxonomy (Section 5.3); 203 synthesized long-document instances over 42 database schemas from BIRD and Spider in seven domain groups; identical prompts, greedy decoding at temperature 0; cells aligned by global maximum-weight tuple matching, a cell matching on exact numeric equality or at least 90 percent string similarity). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.

Top models

#ModelScore
1GPT-5.482.97
2Gemini 2.5 Pro67.26
3Claude Opus 4.664.82
4GPT-4o60.18
5Qwen 3 Max58.57
6Llama 3.1 70B Instruct53.85
7DeepSeek V4 Flash53.83
8Qwen 2.5 14B Instruct47.65

Interactive version: theaggregate.ai/benchmark?slug=doc2db-bench-multi-entity-binding · How It Works · Data refreshed daily, snapshot 2026-09-29.