Gemma 4 31B (High): benchmark results
Provider: Google. Released 2026-04-02. Access: Open.
Aggregate Rating 1671 ± 24, rank #452 of 2170 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ObviousBench | 100 | Answer pass³ (%) | 96.8 |
| Korean CSAT 2026 (Easy Mode) - English | 100 | Points (out of 100) | 90.7 |
| Korean CSAT 2026 (Easy Mode) - Mathematics | 98.67 | Points (out of 100) | 72 |
| PetriBench - Boundedness (Hard) | 52 | Exact-Match Accuracy (%) | 71.4 |
| Korean CSAT 2026 (Easy Mode) - Korean | 97 | Points (out of 100) | 71 |
| Korean CSAT 2026 (Easy Mode) - Total | 419.67 | Points (out of 450) | 66.3 |
| Korean CSAT 2026 (Easy Mode) - Life Science I | 40 | Points (out of 50) | 63.3 |
| Korean CSAT 2026 (Easy Mode) - Society and Culture | 40 | Points (out of 50) | 60 |
| Korean CSAT 2026 (Easy Mode) - Chemistry I | 43 | Points (out of 50) | 54 |
| LLMConfBench | 0.27 | Conformer Ranking Accuracy (mean reliable-pair Kendall tau) | 53.7 |
| PetriBench - L4-Liveness (Hard) | 54 | Exact-Match Accuracy (%) | 51.4 |
| PetriBench - Deadlock (Hard) | 51 | Exact-Match Accuracy (%) | 42.9 |
Interactive version: theaggregate.ai/model?slug=gemma-4-31b-high · How It Works · Data refreshed daily, snapshot 2026-11-05.