gemma-4-31B-it-20260402: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1718 ± 37, rank #278 of 10916 rated models, from 150 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Evals for Every Language - Ganda (lg) | 66.47 | Average Score (%) | 95.3 |
| Evals for Every Language - Lingala (ln) | 67.02 | Average Score (%) | 92.3 |
| Evals for Every Language - Kinyarwanda (rw) | 63.96 | Average Score (%) | 92.2 |
| Evals for Every Language - ARC | 97.16 | Average Score (%) | 91.2 |
| Evals for Every Language - MMLU | 97.81 | Average Score (%) | 91.2 |
| Evals for Every Language - Kannada (kn) | 74.59 | Average Score (%) | 90.6 |
| Evals for Every Language - Punjabi (pa) | 79.2 | Average Score (%) | 90.6 |
| Evals for Every Language - Ewe (ee) | 59.27 | Average Score (%) | 90.3 |
| Evals for Every Language - MGSM | 87.1 | Average Score (%) | 89.9 |
| Evals for Every Language - Danish (da) | 76.79 | Average Score (%) | 89.4 |
| Evals for Every Language - Polish (pl) | 74.56 | Average Score (%) | 89.4 |
| Evals for Every Language - Balinese (ban) | 68.71 | Average Score (%) | 89.1 |
Interactive version: theaggregate.ai/model?slug=gemma-4-31b-it-20260402 · How It Works · Data refreshed daily, snapshot 2026-11-08.