LA Leaderboard - PAWS Catalan: leaderboard
Adversarial paraphrase identification (PAWS) 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 | Mistral 7B Instruct (v0.3) | 72.45 | #1255 |
| 2 | Qwen3-8B-Base | 70.95 | #626 |
| 3 | Yi-1.5-9B | 70.5 | #1215 |
| 4 | Llama 3.1 8B Instruct | 68.95 | #1017 |
| 5 | Gemma 2 9B (IT) | 68.35 | #795 |
| 6 | OLMo-2-1124-7B-Instruct | 68.25 | #1343 |
| 7 | Yi-1.5-9B Chat | 68.05 | #1032 |
| 8 | Llama 3.2 3B Instruct | 67.75 | #1327 |
| 9 | aya-expanse-8B | 67.55 | #914 |
| 10 | Qwen 2.5 3B | 67.35 | #1177 |
| 11 | EuroLLM-9B | 67.15 | #1238 |
| 12 | Gemma 2 9B | 67.1 | #924 |
| 13 | SmolLM3-3B-Base | 67.1 | #1260 |
| 14 | EuroLLM-9B-Instruct | 66.7 | #1047 |
| 15 | Qwen 2.5 7B | 66.65 | #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-paws-catalan · How It Works · Data refreshed daily, snapshot 2026-11-05.