Gemma 4 26B A4B (IT) (Non-reasoning): benchmark results
Provider: Google. Released 2026-04-02. Access: Open.
Aggregate Rating 1598 ± 21, rank #731 of 2170 rated models, from 49 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DAR-MGSM - Average - Localized | 85.6 | Accuracy (%) | 90 |
| DAR-MGSM - Average - Translated | 85.8 | Accuracy (%) | 90 |
| DAR-MGSM - Egyptian - Translated | 88 | Accuracy (%) | 90 |
| DAR-MGSM - Emirati - Translated | 91.6 | Accuracy (%) | 90 |
| DAR-MGSM - MSA - Localized | 86 | Accuracy (%) | 90 |
| DAR-MGSM - Moroccan - Translated | 80.4 | Accuracy (%) | 90 |
| DAR-MGSM - Saudi - Localized | 80 | Accuracy (%) | 90 |
| DAR-MGSM - Saudi - Translated | 78.4 | Accuracy (%) | 90 |
| DAR-MGSM - Egyptian - Localized | 87.6 | Accuracy (%) | 85 |
| DAR-MGSM - Emirati - Localized | 90 | Accuracy (%) | 85 |
| DAR-MGSM - Jordanian - Localized | 90.4 | Accuracy (%) | 85 |
| MERA v2 - Characters | 58.8 | Score (%) | 80.4 |
Interactive version: theaggregate.ai/model?slug=gemma-4-26b-a4b-it-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-05.