LA Leaderboard - WiC Basque: leaderboard
Word-in-Context in Basque: whether a word has the same sense in two sentences (WiCeu). 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 2 9B (IT) | 54 | #795 |
| 2 | salamandra-7B-instruct | 52.71 | #1392 |
| 3 | Gemma 2 9B | 52.43 | #924 |
| 4 | Gemma 3 4B (IT) | 52.14 | #997 |
| 5 | Llama 3.1 8B Instruct | 51.07 | #1017 |
| 6 | salamandra-7B | 50.93 | #1446 |
| 7 | Phi-1.5 | 50.29 | #1526 |
| 8 | Qwen3-8B-Base | 50.21 | #626 |
| 9 | gemma-3-1B-pt Base | 50.14 | #1519 |
| 10 | Qwen 2.5 1.5B | 50.14 | #1412 |
| 11 | Yi-1.5-9B | 50.07 | #1215 |
| 12 | occiglot-7B-eu5-instruct | 50.07 | |
| 13 | Qwen 2.5 7B Instruct | 50 | #863 |
| 14 | Mistral 7B Instruct (v0.3) | 50 | #1255 |
| 15 | Qwen 3 4B | 50 | #834 |
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-wic-basque · How It Works · Data refreshed daily, snapshot 2026-11-05.