LA Leaderboard - ParafrasesGL: leaderboard
Paraphrase identification on Galician sentence pairs (Parafrases_gl). 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 | aya-expanse-8B | 58.5 | #914 |
| 2 | Yi-1.5-9B Chat | 56.12 | #1032 |
| 3 | Gemma 3 4B (IT) | 55.1 | #997 |
| 4 | Qwen 2.5 7B | 55.1 | #895 |
| 5 | Gemma 2 2B (IT) | 54.42 | #1328 |
| 6 | Qwen 3 8B | 54.08 | #678 |
| 7 | Llama 3.2 3B Instruct | 54.08 | #1327 |
| 8 | Qwen 2.5 3B Instruct | 54.08 | #1159 |
| 9 | OLMo-2-1124-7B-Instruct | 54.08 | #1343 |
| 10 | Mistral 7B Instruct (v0.3) | 53.74 | #1255 |
| 11 | EuroLLM-9B | 53.74 | #1238 |
| 12 | occiglot-7B-eu5 | 53.74 | |
| 13 | salamandra-2B | 53.74 | #1520 |
| 14 | Qwen 2.5 3B | 53.4 | #1177 |
| 15 | Llama 3.1 8B Instruct | 53.06 | #1017 |
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-parafrasesgl · How It Works · Data refreshed daily, snapshot 2026-11-05.