LA Leaderboard - TELEIA: leaderboard
TELEIA Spanish language exam questions. 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 | Gemma 3 4B (IT) | 66.67 | #997 |
| 2 | aya-expanse-8B | 66.67 | #914 |
| 3 | Qwen3-8B-Base | 61.9 | #626 |
| 4 | Llama 3.2 3B Instruct | 57.14 | #1327 |
| 5 | Gemma 2 9B (IT) | 57.14 | #795 |
| 6 | Gemma 2 9B | 57.14 | #924 |
| 7 | SmolLM3-3B-Base | 57.14 | #1260 |
| 8 | gemma-3-4B-pt Base | 52.38 | #1110 |
| 9 | Qwen 2.5 3B | 52.38 | #1177 |
| 10 | Teuken-7B-Instruct-v0.4 | 52.38 | #1408 |
| 11 | Llama 3.1 8B Instruct | 47.62 | #1017 |
| 12 | Mistral 7B Instruct (v0.3) | 47.62 | #1255 |
| 13 | Qwen 2.5 1.5B Instruct | 47.62 | #1393 |
| 14 | Qwen 2.5 3B Instruct | 47.62 | #1159 |
| 15 | Llama 3.2 3B | 47.62 | #1376 |
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-teleia · How It Works · Data refreshed daily, snapshot 2026-11-05.