LA Leaderboard - XNLI Spanish: leaderboard
XNLI natural language inference in Spanish (SpanishBench). 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 | Gemma 2 9B | 53.17 | #924 |
| 2 | occiglot-7B-eu5 | 52.61 | |
| 3 | Llama 3.1 8B | 52.57 | #1130 |
| 4 | occiglot-7B-eu5-instruct | 51.89 | |
| 5 | EuroLLM-9B | 51.81 | #1238 |
| 6 | Gemma 2 2B | 51.81 | #1466 |
| 7 | salamandra-7B | 50.76 | #1446 |
| 8 | Mistral-7B-v0.3 | 50.32 | #1321 |
| 9 | Mistral 7B Instruct (v0.3) | 50.28 | #1255 |
| 10 | Llama 3.1 8B Instruct | 50.2 | #1017 |
| 11 | Qwen 2.5 7B | 50.2 | #895 |
| 12 | EuroLLM-9B-Instruct | 49.88 | #1047 |
| 13 | Gemma 2 9B (IT) | 49.52 | #795 |
| 14 | salamandra-7B-instruct | 49.52 | #1392 |
| 15 | Teuken-7B-Instruct-v0.4 | 49.4 | #1408 |
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-spanish · How It Works · Data refreshed daily, snapshot 2026-11-05.