LA Leaderboard - xStoryCloze Catalan: leaderboard
XStoryCloze story-ending selection in Catalan. 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 | salamandra-7B | 72.67 | #1446 |
| 2 | Gemma 2 9B | 71.94 | #924 |
| 3 | EuroLLM-9B | 70.09 | #1238 |
| 4 | Mistral 7B Instruct (v0.3) | 68.3 | #1255 |
| 5 | Llama 3.1 8B | 68.17 | #1130 |
| 6 | gemma-3-4B-pt Base | 68.1 | #1110 |
| 7 | salamandra-7B-instruct | 67.31 | #1392 |
| 8 | Llama 3.1 8B Instruct | 66.84 | #1017 |
| 9 | EuroLLM-9B-Instruct | 66.45 | #1047 |
| 10 | occiglot-7B-eu5 | 66.45 | |
| 11 | Mistral-7B-v0.3 | 65.92 | #1321 |
| 12 | aya-expanse-8B | 65.59 | #914 |
| 13 | salamandra-2B | 65.52 | #1520 |
| 14 | Gemma 3 4B (IT) | 64.66 | #997 |
| 15 | Qwen 2.5 7B | 64.53 | #895 |
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-xstorycloze-catalan · How It Works · Data refreshed daily, snapshot 2026-11-05.