Tiger-Gemma-9B-v2: benchmark results
TheDrummer's v2 decensor of Gemma 2 9B, redone with a lighter touch to stay closer to stock Gemma behavior. Provider: Other. Released 2024-07-16. Access: Open.
Unified ELO 1521 ± 1, rank #595 of 1392 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 11.97 | Score | 84.3 |
| Open LLM Leaderboard - IFEval | 69.86 | Score | 82.8 |
| Open LLM Leaderboard - MMLU-Pro | 34.58 | Score | 74.8 |
| Open LLM Leaderboard - BBH | 35.47 | Score | 71.7 |
| UGI - Willingness (W/10) | 7 | W/10 Score | 70.4 |
| Open LLM Leaderboard - MATH Level 5 | 18.2 | Score | 67.8 |
| UGI Leaderboard | 33.74 | UGI Score | 46.6 |
| Open LLM Leaderboard - MuSR | 9.32 | Score | 45.1 |
| UGI - Writing | 27.95 | Writing Score | 30.7 |
| UGI - Natural Intelligence | 14.74 | NatInt Score | 17.4 |
Interactive version: theaggregate.ai/model?slug=tiger-gemma-9b-v2 · How It Works · Data refreshed daily, snapshot 2026-09-05.