ProLLM - OpenBook Q&A: leaderboard
ProLLM benchmark measuring LLM accuracy on open-book question answering with access to reference documents.
Metric: Score (%). Source: www.prollm.ai. Saturation forecast: Estimated already saturated. 85 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4 | 94.4 | #181 |
| 2 | QwQ-32B | 91.9 | #421 |
| 3 | Grok 3 Mini | 86.7 | #281 |
| 4 | Kimi K2 | 86.3 | #243 |
| 5 | Command A | 86.3 | #508 |
| 6 | Grok 3 | 85.9 | #304 |
| 7 | GPT-4.1 Mini | 85.1 | #355 |
| 8 | DeepSeek V3 | 85.1 | #321 |
| 9 | Gemini 2.5 Pro | 84.7 | #152 |
| 10 | DeepSeek R1 | 84.7 | #251 |
| 11 | GPT-4.1 | 82.7 | #248 |
| 12 | O1 Preview | 81 | #193 |
| 13 | Gemini 2.5 Flash | 80.6 | #245 |
| 14 | O1 | 80.6 | #183 |
| 15 | Qwen 2.5 72B Instruct | 80.2 | #454 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=prollm-openbook-q-a · How It Works · Data refreshed daily, snapshot 2026-10-20.