LA Leaderboard - XNLI Galician: leaderboard
The "XNLI Galician" subtask of LA Leaderboard. Iberian and Basque language evaluation: tests LLMs on COPA, PiQA, Belebele, and other NLU tasks in Spanish, Catalan, Basque, and Galician.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | salamandra-7B-instruct | 50.23 | #1411 |
| 2 | Gemma 2 9B | 50.11 | #958 |
| 3 | aya-expanse-8B | 49.77 | #931 |
| 4 | Gemma 2 9B (IT) | 49.51 | #788 |
| 5 | Teuken-7B-Instruct-v0.4 | 48.97 | #1430 |
| 6 | salamandra-7B | 48.81 | #1461 |
| 7 | EuroLLM-9B | 48.63 | #1254 |
| 8 | Qwen3-8B-Base | 48.61 | #626 |
| 9 | EuroLLM-9B-Instruct | 48.41 | #1054 |
| 10 | occiglot-7B-eu5-instruct | 48.35 | |
| 11 | Llama 3.1 8B Instruct | 48.17 | #1035 |
| 12 | occiglot-7B-eu5 | 47.67 | |
| 13 | Llama 3.1 8B | 47.55 | #1151 |
| 14 | gemma-3-4B-pt Base | 46.98 | #1134 |
| 15 | salamandra-2B-instruct | 46.84 |
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-xnli-galician · How It Works · Data refreshed daily, snapshot 2026-10-20.