XHotpotQA (Gold Supporting Facts): leaderboard
Metric: Answer F1 (%; one-shot reader given the question and the annotated supporting sentences of the audited V1 validation set, 7,405 mixed-language HotpotQA-derived items, legacy Hotpot answer normalization). Source: arxiv.org. Saturation forecast: Estimated already saturated. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 50.82 |
| 2 | Llama 3.1 405B Instruct | 48.32 |
| 3 | GPT-4o Mini | 45.13 |
| 4 | Gemma 2 27B (IT) | 39.54 |
| 5 | Qwen 2 72B Instruct | 37.95 |
| 6 | aya-23-35B | 36.92 |
| 7 | Gemma 2 9B (IT) | 35.32 |
| 8 | Llama 3.1 8B Instruct | 32.12 |
| 9 | Qwen 2 7B Instruct | 28.12 |
| 10 | aya-23-8B | 22.29 |
Interactive version: theaggregate.ai/benchmark?slug=xhotpotqa-gold-supporting-facts · How It Works · Data refreshed daily, snapshot 2026-09-26.