gemma3:27B [Q4_K_M]: benchmark results
Provider: Other.
Aggregate Rating 1573 ± 50, rank #1350 of 10916 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLMStructBench | 0.74 | Composite score (0-1): the mean of micro-averaged key-value | 97.7 |
| LLMStructBench - Document Correctness | 0.52 | Document correctness (0-1): share of tests whose JSON object | 97.7 |
| Turkish MMLU - University Entrance Exam Basic Sciences | 73 | Accuracy (%) | 95.7 |
| Turkish MMLU | 75.06 | Accuracy (%) | 92.3 |
| Turkish MMLU - TUS | 76 | Accuracy (%) | 86.2 |
| Turkish MMLU - Driving Licence Exam | 85 | Accuracy (%) | 84.5 |
| Turkish MMLU - KPSS | 53 | Accuracy (%) | 81 |
| RFMC - English-Russian | 89.13 | COMET-22 (x100; document-level mean over 1,143 medical-domai | 73.3 |
| RFMC - English-Japanese | 89.67 | COMET-22 (x100; document-level mean over 1,143 medical-domai | 63.6 |
| RFMC - English-Chinese | 89.34 | COMET-22 (x100; document-level mean over 1,143 medical-domai | 50 |
| RFMC - English-German | 88.05 | COMET-22 (x100; document-level mean over 1,143 medical-domai | 50 |
Interactive version: theaggregate.ai/model?slug=gemma3-27b-q4-k-m · How It Works · Data refreshed daily, snapshot 2026-11-08.