EntLORE - Explicit Lookup (GraphRAG): leaderboard

Metric: Answer accuracy (%; L1, 469 questions answered by a stated fact; corpus-induced GraphRAG: an entity graph and community reports induced offline from the released corpus; 907 questions over 2,341 anonymized enterprise documents; programmatic scoring for entity, set, count and ordered answers, atomic-claim entailment judged by Claude Opus 4.8 for free-form answers). Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.

Top models

#ModelScore
1GLM-5.268.5
2DeepSeek V4 Flash (Reasoning)67.6
3DeepSeek V4 Pro (Reasoning)66.2
4Claude Sonnet 4.6 (Thinking)65.8
5Qwen 3.5 397B A17B65.6
6Kimi K2.663.4
7GPT-5.456.7
8GPT-5.4 Mini52.2

Interactive version: theaggregate.ai/benchmark?slug=entlore-explicit-lookup-graphrag · How It Works · Data refreshed daily, snapshot 2026-09-29.