SupraBench - Host-Guest Description: leaderboard
Metric: ROUGE-1 F1 x100 (0-100) of open-ended descriptions of candidate hosts or binding guests against expert references (135 questions), zero-shot direct prompt. Source: arxiv.org. Saturation forecast: Around 2034. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 15.2 |
| 2 | GPT-5.4 Mini | 13.7 |
| 3 | Qwen 3.5 27B | 12.2 |
| 4 | Gemini 3 Flash | 11.8 |
| 5 | GPT-5.4 Nano | 10.7 |
| 6 | Llama 3.1 8B Instruct | 9.2 |
| 7 | Qwen 3.5 9B | 4.3 |
Interactive version: theaggregate.ai/benchmark?slug=suprabench-host-guest-description · How It Works · Data refreshed daily, snapshot 2026-09-29.