82.1%
Exact match
Run details
- Date
- 2026-06-11
- Test cases
- 11,759
- Median latency
- 64.35s
- Total cost
- $4.53
- Avg. prompt tokens
- 1,702.9
- Avg. answer tokens
- 502.5
- Avg. reasoning tokens
- —
- Quantization
- bf16
About this benchmark
German MMLU-ProX across 14 subjects; primary metric accuracy. Legacy rows have 11,759 items. Corrected v1.1 rows have 11,737 after a frozen, model-independent manifest excludes 20 duplicated-gold translation collisions, one independently reviewed ambiguous source item, and one independently confirmed notation-translation collapse; the row's item count identifies the definition.
Subject breakdown
| Subject | Score |
|---|---|
| biology | 91.2% |
| business | 87.5% |
| chemistry | 86.7% |
| computer science | 85.9% |
| economics | 86.8% |
| engineering | 77.1% |
| health | 74.1% |
| history | 74.5% |
| law | 59.0% |
| math | 93.5% |
| other | 73.7% |
| philosophy | 74.3% |
| physics | 89.0% |
| psychology | 83.6% |
Provenance
- Canonical model
- google/gemma-4-31b-it
- Run ID
2026-06-11T15-10-23__openai__gemma-4-31b-it-bf16__mmlu_prox_de