salamandra-7B-instruct: benchmark results
Provider: BSC. Released 2024-09-30. Access: Open.
Unified ELO 1342 ± 17, rank #1383 of 1632 rated models, from 356 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LA Leaderboard - OpenBookQA Catalan | 34 | Accuracy (%) | 98.5 |
| LA Leaderboard - TE-ca | 53.19 | Accuracy (%) | 98.5 |
| LA Leaderboard - WiC Basque | 52.71 | Accuracy (%) | 98.5 |
| LA Leaderboard - XNLI Galician | 50.23 | Accuracy (%) | 98.5 |
| LA Leaderboard - EusProficiency Basque | 37.61 | Accuracy (%) | 97.1 |
| LA Leaderboard - EusTrivia Basque | 49.8 | Accuracy (%) | 95.6 |
| LA Leaderboard - OpenBookQA Galician | 31.8 | Accuracy (%) | 95.6 |
| LA Leaderboard - PIQA Catalan | 69.97 | Accuracy (%) | 95.6 |
| LA Leaderboard - SIQA Catalan | 49.54 | Accuracy (%) | 95.6 |
| LA Leaderboard - COPA Catalan | 81.2 | Accuracy (%) | 94.1 |
| LA Leaderboard - BEC2016 Basque | 57.14 | F1 (%) | 92.6 |
| LA Leaderboard - XCOPA Basque | 58.6 | Accuracy (%) | 92.6 |
Interactive version: theaggregate.ai/model?slug=salamandra-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-10-08.