Cydonia-22B-v1.2: benchmark results
TheDrummer's roleplay and creative-writing tune of Mistral Small 22B. Provider: BeaverAI. Released 2024-10-07. Access: Open.
Aggregate Rating 1512 ± 52, rank #2056 of 10870 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 33.05 | Score | 80.7 |
| Open LLM Leaderboard - BBH | 58.09 | Score | 79.2 |
| Open LLM Leaderboard - MMLU-Pro | 41.41 | Score | 75.4 |
| Open LLM Leaderboard - MATH Level 5 | 20.32 | Score | 71.5 |
| Open LLM Leaderboard - IFEval | 56.35 | Score | 68.3 |
| Open LLM Leaderboard - MuSR | 40.22 | Score | 44 |
| StickToYourRole | 56.6 | Cardinal Score | 38.7 |
| Stick To Your Role! | 0.57 | Cardinal Score (-1 to 1) | 36.7 |
| Stick To Your Role! - Rank-Order Stability | 0.49 | Rank-Order Stability (correlation, -1 to 1) | 36.7 |
Interactive version: theaggregate.ai/model?slug=cydonia-22b-v1-2 · How It Works · Data refreshed daily, snapshot 2026-11-12.