LA Leaderboard - PAWS Galician: leaderboard
Adversarial paraphrase identification (PAWS) in Galician. 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 | Qwen3-8B-Base | 68.65 | #626 |
| 2 | Mistral 7B Instruct (v0.3) | 67.85 | #1255 |
| 3 | Gemma 2 9B | 67.3 | #924 |
| 4 | Gemma 2 9B (IT) | 67.2 | #795 |
| 5 | Teuken-7B-Instruct-v0.4 | 66.2 | #1408 |
| 6 | Qwen 2.5 3B | 65.7 | #1177 |
| 7 | OLMo-2-1124-7B-Instruct | 65.7 | #1343 |
| 8 | Llama 3.1 8B Instruct | 65.4 | #1017 |
| 9 | EuroLLM-9B-Instruct | 65.3 | #1047 |
| 10 | SmolLM3-3B | 65 | #1239 |
| 11 | Llama 3.2 3B Instruct | 64.8 | #1327 |
| 12 | gemma-3-4B-pt Base | 64.6 | #1110 |
| 13 | EuroLLM-9B | 64.35 | #1238 |
| 14 | SmolLM3-3B-Base | 64.15 | #1260 |
| 15 | Qwen 2.5 7B Instruct | 64.1 | #863 |
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-galician · How It Works · Data refreshed daily, snapshot 2026-11-05.