DrBencher - Entity Identification: leaderboard
DrBencher (2026) generates questions that need multi-hop entity identification, property retrieval from authoritative sources and domain computation, built answer-first from knowledge-graph chains, and evaluates models closed-book on a human-validated set. This board is the percentage of questions whose hidden target entity the model identifies correctly, extracted from its ENTITY tag and scored by normalized substring matching against gold names, the mean of three runs; higher is better.
Metric: Entity identification accuracy (%): share of questions whose target entity the model identifies, judged by gpt-oss-120b, on the 268 human-validated DrBencher questions (closed-book, default API temperature, mean of three runs, 2% relative tolerance and exact match for history); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 6 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Claude Opus 4.6 | 86.1 | #113 | |
| 2 | GPT-5.2 | 82.1 | #185 | |
| 3 | Gemini 2.5 Flash | 77.4 | #463 | |
| 4 | Llama 4 Maverick | 75.5 | #1110 | |
| 5 | Mistral Small 3.2 | 70.6 | #1641 | |
| 6 | Qwen 3 30B A3B | Thinking | 62.8 | #1156 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=drbencher-entity-identification · How It Works · Data refreshed daily, snapshot 2026-11-21.