SeaEval - Cross-Lingual Consistency - Cross-XQuAD (Zero-Shot) — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 96.83 |
| 2 | Llama 3.1 70B Instruct | 96.16 |
| 3 | Qwen 2 72B Instruct | 96.13 |
| 4 | Llama 3 70B Instruct | 95.92 |
| 5 | Qwen 2.5 14B Instruct | 95.82 |
| 6 | Gemma 2 9B (IT) | 95.67 |
| 7 | Qwen 2.5 7B Instruct | 94.6 |
| 8 | Qwen 2 7B Instruct | 94.18 |
| 9 | SeaLLMs-v3-7B-Chat | 94.03 |
| 10 | Llama 3.1 8B Instruct | 93.8 |
| 11 | Qwen 2.5 3B Instruct | 93.78 |
| 12 | Llama 3 8B Cpt Sea Lionv2.1 Instruct | 93.61 |
| 13 | Llama 3 8B Instruct | 92.1 |
| 14 | Gemma 2 2B (IT) | 91.7 |
| 15 | Qwen 2.5 1.5B Instruct | 89.39 |
Interactive version: theaggregate.ai/benchmark?slug=seaeval-cross-lingual-consistency-cross-xquad-zero-shot · How the rankings work · Data refreshed daily, snapshot 2026-07-22.