RESCUE-BENCH - Support Target Prediction: leaderboard
Metric: MRR (%). Source: arxiv.org. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek V4 Pro | 82.23 |
| 2 | Qwen 3.5 Plus | 81.57 |
| 3 | GPT-4o | 80 |
| 4 | Qwen Plus | 79.21 |
| 5 | DeepSeek V4 Flash | 77.61 |
| 6 | DeepSeek V3.2 | 77.14 |
| 7 | MiniMax-M2.5 | 77.01 |
| 8 | DeepSeek R1 | 76.88 |
| 9 | Qwen 3 Max | 76.28 |
| 10 | Kimi K2.5 | 75.6 |
Interactive version: theaggregate.ai/benchmark?slug=rescue-bench-support-target-prediction · How It Works · Data refreshed daily, snapshot 2026-09-21.