80.7%
Exact match
Run details
- Date
- 2026-08-14
- Test cases
- 11,737
- Median latency
- 2.92s
- Total cost
- $3.44
- Avg. prompt tokens
- 1,718.4
- Avg. answer tokens
- 184.8
- Avg. reasoning tokens
- —
- Quantization
- Unverified
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 | 90.1% |
| business | 83.4% |
| chemistry | 85.9% |
| computer science | 83.6% |
| economics | 85.2% |
| engineering | 71.4% |
| health | 75.1% |
| history | 74.2% |
| law | 58.7% |
| math | 91.5% |
| other | 77.1% |
| philosophy | 75.7% |
| physics | 85.1% |
| psychology | 83.9% |
Provenance
- Canonical model
- deepseek/deepseek-v4-pro-0813
- Run ID
2026-08-14T11-20-28__openrouter__deepseek__deepseek-v4-pro-0813__mmlu_prox_de