Gemma 4 4B: benchmark results
Provider: Google. Access: API.
Aggregate Rating 1525 ± 43, rank #2006 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UnrestSent200K - Few-Shot | 65 | F1 (%) | 85.7 |
| UnrestSent200K - Few-Shot Accuracy | 66.2 | Accuracy (%) | 78.6 |
| ECGQuest | 62.9 | Accuracy (%; 1,050 held-out true/false questions on ECG know | 61.4 |
| UnrestSent200K - Zero-Shot | 49 | F1 (%) | 50 |
| UnrestSent200K - Zero-Shot Accuracy | 52 | Accuracy (%) | 46.7 |
| DGEval - Regulatory Recall | 3.1 | Accuracy (%) | 17.6 |
| DGEval - IMDG Code MCQ | 47.3 | Accuracy (%) | 14.7 |
| CardioLens | 42.19 | F1 (Random) (self-reported) | 13 |
| TempGlitch | 50 | Acc. (1 FPS) (self-reported) | 11.1 |
| SWE-Flux | 17.29 | Accuracy (%; execution-reasoning questions answered correctl | 0 |
Interactive version: theaggregate.ai/model?slug=gemma-4-4b · How It Works · Data refreshed daily, snapshot 2026-11-08.