LA Leaderboard - WNLI Basque: leaderboard
Basque translation of WNLI, Winograd-schema pronoun resolution posed as entailment. From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; accuracy (%), higher is better. Each model result file is a separate run.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 8B | 63.38 | #678 |
| 2 | Llama 3.1 8B Instruct | 61.97 | #1017 |
| 3 | Phi-1.5 | 61.97 | #1526 |
| 4 | aya-expanse-8B | 60.56 | #914 |
| 5 | EuroLLM-9B | 59.15 | #1238 |
| 6 | Mistral-7B-v0.3 | 59.15 | #1321 |
| 7 | Llama 3.2 3B Instruct | 57.75 | #1327 |
| 8 | Llama 3.1 8B | 57.75 | #1130 |
| 9 | Yi-1.5-9B | 57.75 | #1215 |
| 10 | salamandra-2B-instruct | 57.75 | |
| 11 | Qwen 3 4B | 56.34 | #834 |
| 12 | Qwen 3 1.7B | 56.34 | #1175 |
| 13 | DeepSeek-R1-Distill-Qwen-7B | 56.34 | #1401 |
| 14 | DeepSeek R1 Distill Qwen 1.5B | 56.34 | #1521 |
| 15 | Gemma 2 2B | 56.34 | #1466 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=la-leaderboard-wnli-basque · How It Works · Data refreshed daily, snapshot 2026-11-05.