Active Evidence-Seeking and Diagnostic Reasoni — leaderboard
Metric: Task 2: Active Seeking Exact Accuracy (self-reported). Source: benchmarklist.com. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 49.4 |
| 2 | Qwen 3 32B | 48.3 |
| 3 | Qwen 3 14B | 45.1 |
| 4 | Qwen 3 8B | 39.5 |
| 5 | Qwen 2.5 32B Instruct | 39.1 |
| 6 | GLM-4 32B | 35.7 |
| 7 | Qwen 2.5 14B Instruct | 34 |
| 8 | Llama 3 8B Instruct | 24.8 |
| 9 | Qwen 2.5 7B Instruct | 23.3 |
| 10 | DeepSeek R1 Distill Qwen 14B | 23.3 |
| 11 | DeepSeek-R1-Distill-Qwen-7B | 6 |
Interactive version: theaggregate.ai/benchmark?slug=active-evidence-seeking-and-diagnostic-reasoni · How the rankings work · Data refreshed daily, snapshot 2026-07-22.