EntLORE - Cross-Source Composition (LLM Wiki): leaderboard
EntLORE (2026) builds an anonymized enterprise world of 2,341 routine documents from an audited truth graph. This board is answer accuracy in percent on its cross-source composition questions (L2), which combine facts stated in several sources, answered through an agent that browses the corpus compiled offline into a navigable wiki. Higher is better.
Metric: Answer accuracy (%; L2, 204 questions composing facts stated in several sources; LLM wiki: the corpus compiled offline into a navigable wiki read through the same agent loop; 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 February 2027. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Flash (Reasoning) | 59.6 | #258 (DeepSeek V4 Flash) |
| 2 | DeepSeek V4 Pro (Reasoning) | 56.9 | #188 (DeepSeek V4 Pro) |
| 3 | GLM-5.2 | 53.2 | #138 |
| 4 | GPT-5.4 | 52.5 | #123 |
| 5 | Claude Sonnet 4.6 (Thinking) | 51.5 | #161 (Claude Sonnet 4.6) |
| 6 | Kimi K2.6 | 49.7 | #178 |
| 7 | Qwen 3.5 397B A17B | 49.3 | #246 |
| 8 | GPT-5.4 Mini | 47.1 | #389 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=entlore-cross-source-composition-llm-wiki · How It Works · Data refreshed daily, snapshot 2026-11-13.