LA Leaderboard - CatCoLA: leaderboard
Grammatical acceptability judgements on Catalan sentences (CatCoLA). From the LA Leaderboard, which evaluates models on Spanish, Catalan, Basque and Galician tasks with the lm-evaluation-harness; Matthews correlation x100 (signed; chance is 0), higher is better. Each model result file is a separate run.
Metric: MCC (x100). Source: huggingface.co. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 2 9B (IT) | 22.32 | #795 |
| 2 | salamandra-7B | 16.94 | #1446 |
| 3 | EuroLLM-9B-Instruct | 16.71 | #1047 |
| 4 | EuroLLM-9B | 16.63 | #1238 |
| 5 | Gemma 2 9B | 16.3 | #924 |
| 6 | Gemma 3 4B (IT) | 14.66 | #997 |
| 7 | Qwen3-8B-Base | 11.19 | #626 |
| 8 | Mistral 7B Instruct (v0.3) | 9.73 | #1255 |
| 9 | Llama 3.1 8B Instruct | 9.49 | #1017 |
| 10 | Qwen 2.5 3B Instruct | 8.89 | #1159 |
| 11 | aya-expanse-8B | 8.29 | #914 |
| 12 | Qwen 2.5 7B Instruct | 6.93 | #863 |
| 13 | SmolLM3-3B | 6.58 | #1239 |
| 14 | Teuken-7B-Instruct-v0.4 | 6.41 | #1408 |
| 15 | OLMo-2-1124-7B-Instruct | 6.26 | #1343 |
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-catcola · How It Works · Data refreshed daily, snapshot 2026-11-05.